Arturo Angulo Sibaja 
 
 
 
 
 

Gross brain morphology of the Loricariinae (Siluriformes: 
Loricariidae): Comparative anatomy, ecological implications and 

phylogenetic analysis 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 
São José do Rio Preto 

2019  

Câmpus de São José do Rio Preto 



 

Arturo Angulo Sibaja 
 
 
 

 
Gross brain morphology of the Loricariinae (Siluriformes: 

Loricariidae): comparative anatomy, ecological implications and 
phylogenetic analysis 

 
 

 
 

Tese apresentada como parte dos requisitos 
para obtenção do título de Doutor em Biologia 
Animal, junto ao Programa de Pós-Graduação 
em Biologia Animal, do Instituto de Biociências, 
Letras e Ciências Exatas da Universidade 
Estadual Paulista “Júlio de Mesquita Filho”, 
Câmpus de São José do Rio Preto. 
 
Financiadora: PAEC OEA-GCUB / MICITT – 
PED-017-2015-1 

 
Orientador: Prof. Dr. Francisco Langeani Neto 

 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

São José do Rio Preto 
2019 



A594g
Angulo, Arturo

    Gross brain morphology of the Loricariinae (Siluriformes:

Loricariidae): Comparative anatomy, ecological implications and

phylogenetic analysis / Arturo Angulo. -- São José do Rio Preto, 2019

    227 f. : il., tabs., fotos + 1 CD-ROM

    Tese (doutorado) - Universidade Estadual Paulista (Unesp), Instituto

de Biociências Letras e Ciências Exatas, São José do Rio Preto

    Orientador: Francisco Langeani

    1. Zoologia. 2. Ecologia. 3. Peixes de água doce. 4. Cascudo

(Peixe). 5. Cérebro. I. Título.

Sistema de geração automática de fichas catalográficas da Unesp. Biblioteca do Instituto de

Biociências Letras e Ciências Exatas, São José do Rio Preto. Dados fornecidos pelo autor(a).

Essa ficha não pode ser modificada.



 

Arturo Angulo Sibaja 
 
 

Gross brain morphology of the Loricariinae (Siluriformes: 
Loricariidae): comparative anatomy, ecological implications and 

phylogenetic analysis 
 

 
Tese apresentada como parte dos requisitos 
para obtenção do título de Doutor em Biologia 
Animal, junto ao Programa de Pós-Graduação 
em Biologia Animal, do Instituto de Biociências, 
Letras e Ciências Exatas da Universidade 
Estadual Paulista “Júlio de Mesquita Filho”, 
Câmpus de São José do Rio Preto. 
 
Financiadora: PAEC OEA-GCUB / MICITT – 
PED-017-2015-1 

 
 

Comissão Examinadora 
 
Prof. Dr. Francisco Langeani Neto 
UNESP – Câmpus de São José do Rio Preto 
Orientador 
 
Prof. Dr. Fabio Müller Pupo 
USP – Museu de Zoologia, Coleção de Peixes  
 
Prof. Dr. Oscar Akio Shibatta 
UEL – Departamento de Biologia Animal e Vegetal 
 
Prof. Dr. Raphaël Covain 
Muséum d’Histoire Naturelle, Genève 
 
Prof. Dr. Vitor Abrahão 
USP – Museu de Zoologia, Coleção de Peixes  
 
Profª. Drª. Lúcia Rapp Py-Daniel 
INPA – Coleção de Peixes 
 
 

São José do Rio Preto 
12 de Fevereiro 2019 



 

 
ACKNOWLEDGEMENTS 

 
 
To my family, friends, colleagues and teachers (in Brazil and Costa Rica), who 
supported me and provided me all the necessary tools to successfully complete this 
academic phase of my life. To Carolina Méndez for her constant support, patience 
and love in the distance. 
 
To Mark Sabaj, Mariangeles Arce (ANSP), Jonathan Armbruster, Dave Werneke 
(AUM), Armando Ortega (IMCN), Lucia Rapp Py-Daniel (INPA-ICT), Claudio de 
Oliveira (LBP), Carlos Lucena, Alejandro Londoño (MCP), Raphael Covain (MHNG), 
Hernan Ortega (MUSM), Aléssio Datovo, Osvaldo Oyakawa (MZUSP), Carla 
Pavaneli, Sandra Regina (NUP), Flávio Bockmann, Hertz Santos (LIRP), Myrna 
López, Ana Ramírez (UCR), Luiz Malabarba, Juliana Wingert (UFRGS), Mónica 
Rodriguez (UFV), Jeff Williams (USNM), Francisco Severo, Fernando Carvalho 
(ZUFMS) and Marcelo Loureiro (ZVCP) for the loan/donation of specimens for 
dissection.  
 
To the authorities and the administrative staff of the IMCN, MZUSP, LIRP, UCR and 
DZSJRP (IBILCE – UNESP) for the assistance provided and for the use of facilities 
during my work and visits to the respective collections.  
 
To the Costa Rican Ministerio de Ciencia, Tecnología y Telecomunicaciones 
(MICITT; PED-017-2015-1) and the Partnerships Program for Education and Training 
of the Organization of American States in conjunction with the Grupo Coimbra de 
Universidades Brasileiras (PAEC OEA-GCUB, 2014) for financial support. 
 
To Danilo de Oliveira for reviewing and correcting the format of the references.  



 

RESUMO 
 
 

A família Loricariidae é a mais diversa dentro do ordem Siluriformes, contendo cerca 

de 980 espécies válidas. Os membros da família são facilmente reconhecidos por 

possuírem corpos cobertos por placas dérmicas ossificadas, dentes tegumentares 

abundantes e um disco oral ventral que facilita a fixação á superfície e a 

alimentação. A subfamília Loricariinae é a segunda mais diversa dentro de 

Loricariidae, contendo 31 gêneros e cerca de 243 espécies válidas. Os membros da 

subfamília são facilmente reconhecidos por ter um pedúnculo caudal alongado e 

deprimido e por a falta de uma nadadeira adiposa. As espécies de Loricariinae estão 

amplamente distribuídas pelas principais drenagens da maioria dos rios da América 

Central e do Sul, sendo geralmente encontradas próximas ao substrato e 

apresentando uma extraordinária diversidade morfológica e funcional. 

Tradicionalmente, a maioria dos estudos morfológicos e ecomorfológicos em 

membros da subfamília e a família, seja sob uma abordagem descritiva/comparativa 

e/ou cladística, focalizou o uso de caracteres osteológicos. Este estudo descreve e 

compara a morfologia cerebral superficial dos membros da subfamília e família e 

explora seu significado ou valor ecológico e filogenético. Mais de 300 espécimes 

[incluindo representantes de quase todos os gêneros válidos de Loricariinae, bem 

como de outras sete subfamílias de Loricariidae (Delturinae, Hypoptopomatinae, 

Hypostominae, Lithogeninae, Neoplecostominae, “Otothyrinae” e Rhinelepinae)] e 89 

caracteres foram examinados. Os resultados obtidos sugerem que o tamanho, 

volume e forma relativa das diferentes estruturas cerebrais examinadas varia 

principalmente com o comportamento alimentar e o ambiente preferido. Além disso, 

quando analisados e avaliados em conjunto com hipóteses de relacionamento 

filogenético, estes resultados podem ser considerados como evidência empírica 

apoiando o fato de que a diversidade cerebral nos membros atuais da subfamília e a 

família pode ser o resultado tanto do conservadorismo de nicho filogenético quanto 

da radiação adaptativa repetida. Finalmente, este trabalho fornece dados adicionais 

a serem considerados na árvore evolutiva da subfamília e a família, destacando a 

necessidade de mais estudos integrando informações de diferentes sistemas 

anatômicos em um contexto filogenético. Além disso, este trabalho representa uma 

base para futuras investigações sobre as relações entre a anatomia do cérebro e a 

ecologia dos membros da subfamília, família e ordem em um contexto filogenético. 



 

 

Palavras-chave: Cascudos. Ecomorfologia. Filogenia. Neuroanatomia. Ontogenia.  



 

ABSTRACT 
 

 
The family Loricariidae is the most species-rich within the Siluriformes, containing 

about 980 valid species. Members of the family are easily recognized by having 

bodies covered by ossified dermal plates, abundant integumentary teeth and a 

ventral oral disk that facilitates surface attachment and feeding. The subfamily 

Loricariinae is the second most species rich within the Loricariidae containing 31 

genera and about 243 valid species. Members of the subfamily are easily recognized 

by having an elongate and depressed caudal peduncle and by lacking of adipose fin. 

Species of the Loricariinae are widely distributed throughout the Central and South 

American major river drainages, being usually found near the substrate and showing 

an extraordinary morphological and functional diversity. Traditionally, most of the 

morphological, and ecomorphological, studies on members of the subfamily and the 

family, either under a descriptive/comparative and/or cladistic approach, have 

focused on the use of osteological characters. This study describes and compares 

the gross brain morphology of the members of the subfamily and the family and 

explores their ecological and phylogenetic significance. More than 300 specimens 

[including representatives of almost all valid genera of the Loricariinae, as well as of 

other seven subfamilies of the Loricariidae (Delturinae, Hypoptopomatinae, 

Hypostominae, Lithogeninae, Neoplecostominae, “Otothyrinae” and Rhinelepinae)] 

and 89 characters were examined. The results obtained suggest that the relative 

size, volume and shape of the different brain structures examined varies mostly with 

feeding behaviour and preferred environment. Furthermore, when analyzed and 

evaluated in conjunction with hypotheses of phylogenetic relationships, these results 

can be considered as empirical evidence supporting the fact that brain diversity in 

current members of the subfamily, and the family in general terms, could be the 

result of both phylogenetic niche conservatism and repeated adaptive radiation. 

Finally, this work provides considerable additional data to the evolutionary tree of the 

subfamily and the family, highlighting the needing of further studies integrating 

information from different anatomical systems in a phylogenetic context. Moreover, 

this work could be a basis for further investigations on the relationships between 

brain anatomy and the ecology of members of the subfamily, family and order in a 

phylogenetic frame. 

 



 

Keywords: Ecomorphology. Neuroanatomy. Ontogeny. Phylogeny. Suckermouth 

armoured catfishes. 

  



 

LIST OF TABLES, FIGURES AND ILLUSTRATIONS 
 
 

Table 1. Body length (SL), mass (We) and brain measurements (28) of examined 
specimens of Rineloricaria heteroptera (n=42); absolute values and percentages of 
total brain length, head length or total brain volume, as appropriate. Morphometric 
measurements (-L=length; -W=width, -H=height), numbered from 1 to 28, are 
expressed in mm, volumes (“-V”s) are expressed in mm3 and masses (“-We”s) are 
expressed in mg. Linear measurements refer only to the left lobe or counterpart for 
those bilateral structures; volumes for these bilateral structures were doubled, 
assuming brain symmetry ………………………………………………………...……  27 
 
Table 2. Results of the ANCOVAs evaluating the effect of the body length (SL) and 
the body mass (BW), as the covariates, and the sex, the developmental stage and 
the interaction between them (i.e. sex*developmental stage), as the factors, on the 
scaling of the overall brain length (T-L) and brain mass (T-W) in Rineloricaria 
heteroptera. Significant p values are denoted with an asterisk (*) ……………  42 
 
Table 3. Slopes or "allometric coeffients" (a) and intercepts or "allometric 
components" (b), with their respectives ± standard error of mean (SEM) and 95% 
bootstrapped confidence intervals (CIV; N=1999), and correlation values (r2), with 
their respective associated "p" value, of the OLS linear regressions lines for the eigth 
major brain subdivisions volumes (vs. corrected total brain volumes) in Rineloricaria 
heteroptera .…………..…………………………………...…………………………….  42 
 
Table 4. Results of the ANCOVAs evaluating the effect of the corrected total brain 
volume, as the covariate, and the sex, the developmental stage and the interaction 
between them (i.e. sex*developmental stage), as the factors, on the scaling of each 
brain subdivision volume in Rineloricaria heteroptera. Significant p values are 
denoted with an asterisk (*) ……………………..……………………………….  43 
 
Table 5. Results for the first four components (PCs) of the PCA of the relative 
volume of eight brain subdivisions in Rineloricaria heteroptera …………..….  44 
 
Table 6. Results of the PERMANOVA test for statistical significance between groups 
of the Loricariidae, based on the PCoA using a total of 88 binary or multistate, non-
ordered characters describing its gross brain morphology. Pairwise comparisons; F 
and Bonferroni-corrected p (Bp) values. Abbreviations: TG=Taxonomic group; 
De=Delturinae; Hp=Hypoptopomatinae; Hy=Hypostominae; LF=Loricariinae, 
Loricariini, Farlowellina; LH=Loricariinae, Harttiini; Li=Lithogeninae; LL=Loricariinae, 
Loricariini, Loricariina; Ne=Neoplecostominae; and Rh=Rhinelepinae. Bold numbers 
denotes a significant difference between groups ……………………………...  45 
 
Figure 1. Rineloricaria heteroptera; Machado River basin, Madeira River drainage, 
Rondonia, Brazil; (A) entire specimen (DZSJRP 16730; 75.25 mm SL; female); 
dorsal (above), lateral (centre) and ventral (below) views; (B, C) ventral detail of head 
of sexually dimorphic male (B; DZSJRP 14771, 80.15 mm SL) and female (C; 
DZSJRP 17429, 81.11 mm SL); note the presence (in the male) of shorted, thickened 
an curved pectoral-fin spines and numerous small odontodes along the sides of the 
head and the pectoral-fin spines ...........................................................................  31 



 

 
Figure 2. Topographic brain anatomy of Rineloricaria heteroptera (DZSJPR 014771), 
114.42 mm SL; Machado River basin, Madeira River drainage, Rondonia, Brazil; 
(above) dorsal, (centre) lateral (left side) and (below) ventral views. Pointed lines 
indicate the anterior and posterior limits of the brain ...........................................  36 
 
Figure 3. Intraspecific variation on the brain of Rineloricaria heteroptera; Machado 
River basin, Madeira River drainage, Rondonia, Brazil. (A, B, C) DZSJPR 016730, 
38.75 mm SL; (D) dorsal, (E) lateral (left side) and (F) ventral views. (D, E, F) 
DZSJPR 014771, 54.31 mm SL; (D) dorsal, (E) lateral (left side) and (F) ventral 
views. (G, H, I) DZSJPR 014549, 97.29 mm SL; (G) dorsal, (H) lateral (left side) and 
(I) ventral views. Scale bar=2 mm ...........................................................................  39 
 
Figure 4. Regression analysis (OLS) of the LOG brain size (length) on the LOG body 
size (length) in Rineloricaria heteroptera. Developmental stages are denoted with 
different color symbols (Green=early juveniles, i.e. <60.0 mm SL; Blue=late juveniles, 
i.e. 60.1–80.0 mm SL; and Red=adults, i.e. >80.0 mm SL); females are represented 
by closed symbols, males are represented by open symbols ................................  41 
 
Figure 5. Regression analysis (OLS) of the LOG brain mass on the LOG body mass 
in Rineloricaria heteroptera. Developmental stages are denoted with different color 
symbols (Green=early juveniles, Blue=late juveniles and Red=adults); females are 
represented by closed symbols, males are represented by open symbols ..........  47 
 
Figure 6. Allometric scaling relationships (volume vs. volume OLS regression lines) 
for the eigth major brain subdivisions in Rineloricaria heteroptera. Developmental 
stages are denoted with different color symbols (Green=early juveniles, Blue=late 
juveniles and Red=adults); females are represented by closed symbols, males are 
represented by open symbols .................................................................................  48 
 
Figure 7. Scatterplot of the PCA (PC1 vs. PC2), representing the major changes in 
the composition of the brain along development in Rineloricaria heteroptera. 
Developmental stages are denoted with different color symbols (Green=early 
juveniles, i.e. <60.0 mm SL; Blue=late juveniles, i.e. 60.1–80.0 mm SL; and 
Red=adults, i.e. >80.0 mm SL); females are represented by closed symbols, males 
are represented by open symbols ...........................................................................  52 
 
Figure 8. Topographic brain anatomy of a generalized member of the Loricariinae 
(Loricaria cataphracta, type species of the family; MZUSP 014106; 114.70 mm SL); 
(A) dorsal, (B) lateral (left side) and (C) ventral views. Pointed lines indicate the 
anterior and posterior limits of the brain. Scale bar=2 mm ................................  69 
 
Figure 9. Brain of selected species of the Loricariinae, dorsal view; (A) Farlowella 
oxyrryncha, DZSJRP 014920, female, 124.00 mm SL; (B) Fonchiiloricaria nanodon, 
MUSM 032153, male, 167.98 mm SL; (C) Harttia novalimensis, DZSJRP 001644, 
male, 102.53 mm SL; (D) Hemiodontichthys acipenserinus, INPA-ICT 032082, male, 
92.34 mm SL; (E) Loricariichthys anus, DZSJRP 010987, female, 158.00 mm SL; (F) 
Ricola macrops, ZVCP 014035, male, 221.00 mm SL; (G) Rineloricaria cubataonis, 
DZSJRP 003269, female, 104.53 mm SL; (H) Sturisomatichthys leightoni, ANSP 
084177, male, 118.26 mm SL. Scale bar=2 mm .....................................................  70 



 

 
Figure 10. Brain of selected species of the Loricariinae, lateral view; (A) Farlowella 
oxyrryncha, DZSJRP 014920, female, 124.00 mm SL; (B) Fonchiiloricaria nanodon, 
MUSM 032153, male, 167.98 mm SL; (C) Harttia novalimensis, DZSJRP 001644, 
male, 102.53 mm SL; (D) Hemiodontichthys acipenserinus, INPA-ICT 032082, male, 
92.34 mm SL; (E) Loricariichthys anus, DZSJRP 010987, female, 158.00 mm SL; (F) 
Ricola macrops, ZVCP 014035, male, 221.00 mm SL; (G) Rineloricaria cubataonis, 
DZSJRP 003269, female, 104.53 mm SL; (H) Sturisomatichthys leightoni, ANSP 
084177, male, 118.26 mm SL. Scale bar=2 mm  ...........................................  74 
 
Figure 11. Brain of selected species of the Loricariinae, ventral view; (A) Farlowella 
oxyrryncha, DZSJRP 014920, female, 124.00 mm SL; (B) Fonchiiloricaria nanodon, 
MUSM 032153, male, 167.98 mm SL; (C) Harttia novalimensis, DZSJRP 001644, 
male, 102.53 mm SL; (D) Hemiodontichthys acipenserinus, INPA-ICT 032082, male, 
92.34 mm SL; (E) Loricariichthys anus, DZSJRP 010987, female, 158.00 mm SL; (F) 
Ricola macrops, ZVCP 014035, male, 221.00 mm SL; (G) Rineloricaria cubataonis, 
DZSJRP 003269, female, 104.53 mm SL; (H) Sturisomatichthys leightoni, ANSP 
084177, male, 118.26 mm SL. Scale bar=2 mm .....................................................  76 
 
Figure 12. Empirical morphospace for members of the Loricariidae, based on the 
first three principal coordinates (PCos) of the PCoA using a total of 88 binary or 
multistate non-ordered characters describing its gross brain morphology. (A) PCo1 
vs. PCo2 empirical morphospace; (B) PCo1 vs. PCo3 empirical morphospace; (C) 
Detail of the space occupied by members of the Loricariinae in the PCo1 vs. PCo2 
empirical morphospace; (D) Detail of the space occupied by members of the 
Hypoptopomatinae sensu lato and the Neoplecostominae in the PCo1 vs. PCo2 
empirical morphospace (see complete species names in Appendix 3). Yellow 
symbols represents members of the Delturinae (De); blue symbols represents 
members of the Hypoptopomatinae (Hp); orange symbols represents members of the 
Hypostominae (Hy); red symbols represents members of the Neoplecostominae 
(Ne); green symbols represents members of the Loricariinae [non-filled symbols 
represents members of the Harttiini (LH); the purple symbol represents the single 
representative of the Lithogeninae (Li); filled symbols represents members of the 
Loricariini (LF=Farlowellina, LL=Loricariina)]; the black symbol represents the single 
representative of the Rhinelepinae …………………………………………..….  84 
 
Figure 13. Strict consensus of ten maximally parsimonious trees (1154 steps; 
CI=0.10; RI=0.66), showing the interrelationships among members of the Loricariinae 
and Loricariidae (gross brain morphology); see part of the consensus tree (i.e., non-
Loricariinae members of the Loricariidae) in Fig. 14. Numbers above branches 
indicate the number of the clade and those below the branches correspond to the 
Bremer support values (in those clades in which no number is indicated the Bremer 
support value was equal or greater than 15) …………………………………...… 110 
 
Figure 14. Part of the strict consensus of ten maximally parsimonious trees (1154 
steps; CI=0.10; RI=0.66), showing the interrelationships among members of the non-
Loricariinae subfamilies of the Loricariidae (gross brain morphology); see the 
complete tree in Fig. 13. Numbers above branches indicate the number of the clade 
and those below the branches correspond to the Bremer support values (in those 



 

clades in which no number is indicated the Bremer support value was equal or 
greater than 15) ….........................................................................................… 112 
 
Figure 15. Strict consensus of ten maximally parsimonious trees (1459 steps; 
CI=0.13; RI=0.71), showing the interrelationships among members of the Loricariidae 
(combined analysis); see part of the consensus tree (i.e., members of the 
Loricariinae) in Fig. 16. Numbers above branches indicate the number of the clade 
and those below the branches correspond to the Bremer support values (in those 
clades in which no number is indicated the Bremer support value was equal or 
greater than 15) …………………………………………………………………...… 116 
 
Figure 16. Part of the strict consensus of ten maximally parsimonious trees (1459 
steps; CI=0.13; RI=0.71), showing the interrelationships among members of the 
Loricariinae (combined analysis); see the complete tree in Fig. 15. Numbers above 
branches indicate the number of the clade and those below the branches correspond 
to the Bremer support values (in those clades in which no number is indicated the 
Bremer support value was equal or greater than 15) ……………………………... 118 
 
Figure 17. Detail of the olfactory organs and olfactory bulbs of selected species of 
the Loricariidae, dorsal view. (A) Delturus brevis, NUP 015446, male, 137.66 mm SL; 
(B) Lamontichthys filamentosus, LBP 000162, male, 184.00 mm SL; (C) 
Loricariichthys anus, DZSJRP 010987; female, 158.00 mm SL; (D) Farlowella 
oxyrryncha, DZSJRP 017180, male, 127.00 mm SL. Scale bar=2 mm ……... 149 
 
Figure 18. Detail of the olfactory organs and olfactory bulbs of selected species of 
the Loricariidae, ventral and lateral views. (A) Farlowella henriquei, MZUSP 014408, 
male, 120.62 mm SL; (B) Scleromystax barbatus, DZSJRP 005726, female, 63.70 
mm SL; (C) Ancistrus sp., DZSJRP 011949, female, 53.80 mm SL; (D) Dentectus 
barbarmatus, ANSP 198883, male, 100.52 mm SL; (E) Delturus carinotus, MCP 
028037, male, 153.10 mm SL. Scale bar=2 mm ………………………………..……. 151 
 
Figure 19. Brain of selected species of the Loricariidae; detail of the olfactory organs, 
olfactory bulbs and the anterior portion of the brain (telencephalon and 
mesencephalon), dorsal and lateral views. (A) Corumbataia cuestae, DZSJRP 
008027, female, 29.30 mm SL; (B) Megalancistrus parananus, DZSJRP 004845, 
female, 129.00 mm SL; (C) Limatulichthys petleyi, MZUSP 073995, male, 94.65 mm 
SL; (D) Loricaria cataphracta, DZSJRP 014499, male, 130.13 mm SL; (E) Otocinclus 
affinis, DZSJRP 007610, female, 26.00 mm SL. Scale bar=2 mm ……………... 152 
 
Figure 20. Brain of selected species of the Loricaroidea; detail of the antero-medial 
portion of the brain (telencephalon, diencephalon and mesencephalon), dorsal and 
lateral views. (A) Kronichthys subteres, MCP 020152, female, 64.10 mm SL; (B) 
Corydoras aeneus, DZSJRP 015193, female, 44.80 mm SL; (C) Hypostomus regani, 
DZSJRP 016049, female, 123.00 mm SL, lateral view; (D) Dasyloricaria latiura, 
USNM 293168, male, 242.00 mm SL; (E) Delturus carinotus, MCP 028037, male, 
153.10 mm SL. Scale bar=2 mm …………………………………………………...… 153 
 
Figure 21. Brain of selected species of the Loricariidae, lateral and ventral views. (A) 
Loricaria cataphracta, MZUSP 014106, male, 114.70 mm SL; (B) Farlowella 
paraguayensis, DZSJRP 018794, female, 99.55 mm SL; (C) Cteniloricaria 



 

platystoma, ANSP 190521, female, 116.71 mm SL; (D) Hemiodontichthys 
acipenserinus, INPA-ICT 032082, male, 92.34 mm SL; (E) Otothyropsis  
marapoama, DZSJRP 014108, female, 27.20 mm SL; (F) Paraloricaria vetula, ZVCP 
14036, male, 217.00 mm SL; (G) Neoplecostomus selenae, DZSJRP 015331, male, 
78.30 mm SL; (H) Neoplecostomus yapo, DZSJRP 013651, female, 74.90 mm SL. 
Scale bar=2 mm …………………………………………………………………...… 154 
 
Figure 22. Brain of selected species of the Loricaroidea, detail of the anterior and 
medial portions of the brain (telencephalon, mesencephalon and corpus cerebelli), 
dorsal and lateral views. (A) Neoplecostomus selenae, DZSJRP 015331, male, 78.30 
mm SL; (B) Neoplecostomus yapo, DZSJRP 013651, female, 74.90 mm SL; (C) 
Rineloricaria cadeae, MCP 009782, male, 123.08 mm SL; (D) Corydoras aeneus, 
DZSJRP 015193, female, 44.80 mm SL; (E) Harttia novalimensis, DZSJRP 011644, 
male, 102.53 mm SL; (F) Proloricaria prolixa, DZSJRP 006312, female, 104.51 mm 
SL. Scale bar=2 mm ……………………………………………………………... 157 
 
Figure 23. Brain of selected species of the Loricariidae, dorsal and lateral views. (A) 
Pseudohemiodon platycephalus, ZUFMS-PIS 000440, female, 136.26 mm SL; (B) 
Loricaria luciae, ZUFMS-PIS 003215, male, 120.33 mm SL; (C) Dekeyseria 
scaphirhynchus, INPA-ICT 036022, male, 85.17 mm SL; (D) Hisonotus insperatus, 
DZSJRP, 018211, male, 23.20 mm SL; (E) Loricariichthys platymetopon, DZSJRP 
004395, female, 241.00 mm SL; (F) Delturus brevis, NUP 015446, male, 137.66 mm 
SL; (G) Hisonotus insperatus, DZSJRP, 014381, female, 32.30 mm SL. Scale bar=2 
mm ……………………………………………………………………………………... 123 
 
Figure 24. Brain of selected species of the Loricariidae, dorsal and lateral views. (A) 
Fonchiiloricaria nanodon, MUSM 032153, female, 166.27 mm SL; (B) Reganella 
depressa, MZUSP 057936, male, 139.97 mm SL; (C) Brochiloricaria macrodon, NUP 
002248, male, 262.00 mm SL; (D) Oxyropsis wrighthiana, MCP 034503, female, 
44.00 mm SL; (E) Furcodontichthys novaesi, MZUSP 057726, male, 126.38, mm SL. 
Scale bar=2 mm ………………………………………………………………...…… 162 
           



 

LIST OF ABBREVIATIONS AND ACRONYMS 
 
 
Ad  Adenohypophysis 

Ce  Corpus cerebelli 

Ch  Chiasma opticum 

Char  Character 

De  Delturinae 

EG  Eminentia granularis 

Hp  Adenohypophysis/Hypoptopomatinae 
Hy  Hypothalamus/Hypostominae 

I  Nervus olfatorius 

II  Nervus opticus 

III  Nervus oculomotorius 

IV  Nervus trochlearis 

IX  Nervus glossopharyngeus 

La  Ofactory lamella 

LF  Loricariinae, Loricarini, Farlowellina 

LH  Lobus inferior hypothalami/Loricariinae, Harttiini 

Li  Lithogeninae 

LL  Loricariinae, Loricarini, Loricarina 

LLA  Nervus linea lateralis anterior 

LLP  Nervus linea lateralis posterior 

LobVII Lobus facialis 

LobX  Lobus vagi 

MO  Alar portion of the medulla oblongata 

Me  Medulla spinalis 

Ne  Neoplecostominae 

OB  Olfactory bulbs 

Of  Olfactory organ 

OT  Optic tectum 

Rh  Rhinelepinae 

TC  Truncus cerebri 

SL  Standard length 

Te  Telencephalon 



 

TG  Taxonomic group 

TL  Torus lateralis 

Tol  Nervus tractus olfactorius 

V  Nervus trigeminus 

VI  Nervus abducens 

VII  Nervus facialis 

VIII  Nervus octavus or vestibulares 

X  Nervus vagus 

  



 

TABLE OF CONTENTS 
 
 
1. GENERAL INTRODUCTION ……………………………………………...  17 
1.1. Aims and dissertation structure …………………………………………...…  19 

 
2. CHAPTER 1: GROSS BRAIN MORPHOLOGY OF Rineloricaria heteroptera 

ISBRÜCKER & NIJSSEN, 1976 AS SPECIES MODEL: A DESCRIPTIVE 
AND QUANTITATIVE APPROACH ……………....…………………...…  20 

2.1. Abstract ……………………………………………………………………...  20 
2.2. Keywords …………………………………………………………………..….  21 
2.3. Introduction …………………………………………………………………..….  22 
2.4. Material and methods ……………………………………………………...  24 
2.4.1. Species of study ......................................................................................  24 
2.4.2. Material examined ……………………………………………………………...  25 
2.4.3. Data acquisition ......................................................................................  25 
2.4.4. Data analyses ..............................................................................................  32 
2.5. Results ………………………………………………………………………….…  34 
2.5.1. Gross brain morphology of Rineloricaria heteroptera (Descriptive

approach) .................................................................................................  34 
2.5.2 Sexual dimorphism and ontogenetic variation in the number of lamellae on the 

olfactory organs and in the total brain size and mass and brain subdivisions 
volumes in Rineloricaria heteroptera (Quantitative approach) .....................  40 

2.6. Discussion ………………………………………………………………….…..  47 
2.6.1 Sexual dimorphism in the brain of Rineloricaria heteroptera and its ecological 

implications …………………………………………………………………...…  47 
2.6.2 Ontogenetic variation in the brain of Rineloricaria heteroptera and its 

ecological implications ……………………………………………………...  50 
 
3. CHAPTER 2: GROSS BRAIN MORPHOLOGY OF THE LORICARIINAE: 

COMPARATIVE ANATOMY AND ECOLOGICAL AND PHYLOGENETIC 
CONSIDERATIONS ………………………………………………….......  58 

3.1 Abstract ...……………………………………………………………………  58 
3.2. Keywords …………………………………………………………...…………  59 
3.3. Introduction …………………………………………………………….........….  60 
3.4. Material and methods ……………………………………………………...  62 
3.4.1. Group of study …………….………………………………………………..  62 
3.4.2. Material examined ……………………………………………………………...  63 
3.4.3. Data acquisition ……………………………………………………………...  64 
3.4.4. Data analyses ……………………………………………………………...  65 
3.5. Results ………………………………………………………………...……  66 
3.5.1. Gross brain morphology of the Loricariinae (Descriptive anatomy) …...…  66 
3.5.2. Gross brain morphological disparity in the Loricariidae (Quantitative

approach) .................................................................................................  80 
3.6. Discussion …………………………………………………………….………..  84 
3.6.1. Comparative anatomy and gross ecological correlations and

implications …………………………………..………………………….....…..  84 
3.6.2. Morphological disparity and phylogenetic considerations …………..….  95 
3.6.3. Additional phylogenetic considerations ………………...……………………  97 
 



 

4. CHAPTER 3: A CONTRIBUTION TO THE PHYLOGENY OF THE 
LORICARIIDAE, WITH EMPHASIS ON THE LORICARIINAE, USING 
MOSTLY GROSS BRAIN MORPHOLOGICAL CHARACTERS ................  99 

4.1 Abstract ...……………………………………………………………………  99 
4.2. Keywords …………………………………………………………...………… 100 
4.3. Introduction …………………………………………………………….........…. 101 
4.4. Material and methods ……………………………………………………... 106 
4.4.1. Taxon sampling ……………………………………………………………... 106 
4.4.2. Character sampling ...................................................................................... 107 
4.4.3. Cladistic methodology and phylogenetic analysis ................................ 108 
4.4.4. Combined analysis …………………………………………………….……….. 108 
4.5. Results ………………………………………………………………...…… 109 
4.5.1. Neuroanatomical analysis …………………………………….………..……… 109 
4.5.2. Combined analysis ………………………………………………….………….. 113 
4.6. Discussion ……………………………………………………………………... 119 
 
5. GENERAL CONCLUSIONS ………………………………………...…… 125 
 
 REFERENCES ……………………………………………………………... 126 
 
 APPENDIXES ………………………………………………….......……… 145 

APPENDIX A. Material examined ………………………………………..……. 145 
 APPENDIX B. List of characters analyzed (gross brain morphology) ……... 149 

APPENDIX C. Data matrix used for the PCoA ………………………..……. 164 
APPENDIX D. Data matrix used for the parsimony analyses …………..…. 175 

 APPENDIX E. List of characters analyzed (osteology and other external 
characters) ................................................................................................. 188 

 APPENDIX F. List of synapomorphies and autapomorphies for each clade and 
terminal taxon, respectively (gross brain morphology) ……………………... 193 

 APPENDIX G. List of synapomorphies and autapomorphies for each clade 
and terminal taxon, respectively (combined analysis) ………………….….. 208 

 
 



	

	

17	

	

1. GENERAL INTRODUCTION 
 
 

The family Loricariidae (suckermouth armoured catfishes, armoured catfishes 

or plecos) is the most species-rich within the Siluriformes (catfishes), containing 

about 980 valid species (FRICKEE et al., 2018a) as well as several undescribed 

forms. Members of the family are easily recognized from other fishes by having 

bodies covered by ossified dermal plates, abundant integumentary teeth (odontodes) 

and a ventral oral disk that facilitates surface attachment and feeding (REIS et al. 

2003; GARG et al., 2010; GEERINCKX et al., 2011; LUJAN et al., 2015). The 

subfamily Loricariinae is the second most species-rich within the Loricariidae 

containing 31 genera and about 243 valid species (FRICKEE et al., 2018a) as well as 

several undescribed forms, corresponding to about 25% of the total known diversity 

within the family. Members of the Loricariinae are easily recognized from other 

loricariids by having elongate and depressed caudal peduncles and by lacking of 

adipose fin, among other distinctive characters (see SCHAEFER, 1987; RAPP PY-

DANIEL, 1997; COVAIN & FISCH-MULLER, 2007); they are widely distributed 

throughout the Central and South American major river drainages, from southern 

Costa Rica to northern Argentina (REIS et al., 2003; COVAIN & FISCH-MULLER, 

2007), showing an extraordinary morphological and functional diversity (see 

SCHAEFER & LAUDER, 1986; COVAIN & FISCH-MULLER, 2007; COVAIN, 2011; 

LUJAN & ARMBRUSTER, 2012; LUJAN et al., 2015). 

 Studies focused on the phylogenetic relationships on members of the 

Loricariinae, and Loricariidae, in general terms, agree, in some measure, with the 

monophyly of the subfamily and in the recognition of two major lineages, the tribes 

Harttiini and Loricariini, whose generic composition varies between authors (see for 

example the contributions of RAPP PY-DANIEL, 1997; MONTOYA-BURGOS et al., 

1998; ARMBRUSTER, 2004; COVAIN & FISCH-MULLER, 2007; COVAIN et al., 

2008, 2010, 2016; ROXO et al., 2019). Despite of such studies, and as noted by the 

same authors, a taxonomic synthesis of the subfamily, and the family, in general 

terms, is still needed in order to provide a foundation for more detailed studies on its 

members, considering that (1) at lower taxonomic levels, morphological–molecular 

discordance and homoplasy is still problematic or has not been appropriately 

evaluated, (2) synapomorphies for many clades at/below the rank of tribe are 

relatively scarce and (3) the disagreement about the monophyly and taxonomic 



	

	

18	

	

validity of some genera (see RAPP PY-DANIEL, 1997; COVAIN & FISCH-MULLER, 

2007; RAPP PY DANNIEL & FICHBERG, 2008; COVAIN, 2011; LONDOÑO-

BURBANO, 2012; COVAIN et al., 2016; ROXO et al., 2019).  

 Most of the morphological, and ecomorphological, studies on members of the 

Loricariinae, and the Loricariidae, in general terms, either under a descriptive, 

comparative (e.g., ISBRÜCKER, 1979; RAPP PY-DANIEL, 1997; ARMBRUSTER, 

2004; COVAIN & FISCH-MULLER, 2007; LUJAN & ARMBRUSTER, 2012) and/or 

cladistic approach (e.g., SCHAEFER, 1987; RAPP PY-DANIEL, 1997; 

ARMBRUSTER, 2004; among others), have focused on the use of osteological 

characters; this could be understandable given (1) their “pretty well-known” utility for 

comparative purposes, (2) their relative easy access and (3) the extensive literature 

available (see for example REGAN, 1904; 1911; LEEGE, 1922; ANGELESCU & 

GNERI, 1949; ALEXANDER, 1964; 1965; CHARDON, 1968; LUNDBERG & BASKIN, 

1969; ISBRÜCKER, 1979; SCHAEFER, 1987; RAPP PY-DANIEL, 1997; COVAIN & 

FISCH-MULLER; 2007). On the other hand, descriptive, comparative and/or cladistic 

studies considering alternative body systems, such as the brain anatomy 

(neuroanatomy), in both members of the Loricariinae and Loricariidae, as well as in 

other teleost fish taxa, in general terms, have been rarely addressed, and/or when 

undertaken, rarely carried on in an organized fashion [see FREIHOFER, 1978; 

HOWES, 1983; RAPP PY-DANIEL, 1997; PUPO, 2011, 2015; DATOVO & VARI, 

2014; ROSA, 2015; PEREIRA & CASTRO, 2016; among others (see below), for 

examples and a more detailed discussion]. This “extensive” exploration of 

osteological characters demonstrated to be “rather efficient” for the delimitation of 

most major clades within the Teleostei; however, as noted by DATOVO & VARI 

(2014) and PEREIRA & CASTRO (2016), among other authors, it also resulted in the 

“relatively minor attention” of alternative, and (possibly) equally or more informative, 

anatomical systems.  

The first investigations on the neuroanatomy of members of the Siluriformes 

date to the end of the 19th century (see PUPO & BRITTO, 2018); one of the first 

attempts was the work of HERRICK & HERRICK (1891) who used members of the 

family Ictaluridae (bullhead catfishes) as subject of study. In the middle of the 20th 

century, several studies on the external morphology of the brain took place, followed 

by many papers published on the general anatomy, physiology, cytoarchitecture, 

hodology and embryology of the brain and the peripheral nervous system of, mainly, 



	

	

19	

	

Neartic species (see KOTRSCHAL et al., 1998; FINGER, 2000 for an overview). On 

the other hand, studies on Neotropical fishes, and members of the Siluriformes, 

specifically, are relatively scarce and have focused only in a few species or 

supraspecific groups; e.g., members of the Callichthyidae (PUPO, 2011; PUPO & 

BRITO, 2018), Heptapteridae (TRAJANO, 1994; ABRAHÃO et al., 2018a), 

Loricariidae (ROSA et al., 2014; ROSA, 2015; ANGULO & LANGEANI, 2017; 

CHAMON et al., 2018) and Pseudopimelodidae (ABRAHÃO & SHIBATTA, 2015; 

ABRAHÃO et al., 2018b). Furthermore, in these essentially descriptive (mainly 

anatomical) contributions, the potential ecological (and phylogenetic) inferences and 

correlations have been, in most cases, only superficially addressed (see ROSA, 

2015; PEREIRA & CASTRO, 2016; ANGULO & LANGEANI, 2017; ABRAHÃO et al., 

2018a). 

 

1.1. Aims and dissertation structure  
 

In accordance with the above, the main objective of this work is to describe 

the general gross morphology of the brain of the armoured catfish subfamily 

Loricariinae, as a baseline for comparative anatomical, taxonomic, phylogenetic and 

ecological studies. Moreover, four specific objectives, corresponding each one of 

them (in general terms) with one or more of the three chapters presented below, 

were raised. These four specific objectives are: (1) to explore and evaluate 

(quantitatively and qualitatively) the sexual and ontogenetic (post-larvae) intraspecific 

variation in the gross brain morphology of members of the Loricariinae (chapter one); 

(2) to explore andevaluate and describe (quantitatively and qualitatively) the 

intergeneric (considering the major suprageneric taxonomic divisions) variation in the 

gross brain morphology in members of the Loricariinae, also comparing it with some 

external taxa within the Loricariidae, Siluriformes and Ostariophysi (chapter two); (3) 

to explore and evaluate and discuss the possible correlations between the general 

brain patterns observed and the sensory and behavioral ecology of the species and 

supraspecific groups (chapters one and two); and (4) to evaluate the phylogenetic 

significance of the neuroanatomy, in members of the Loricariinae and the 

Loricariidae, following a cladistic approach (chapter three). 

  

  



	

	

20	

	

2. CHAPTER ONE: 
 
 

GROSS BRAIN MORPHOLOGY OF Rineloricaria heteroptera ISBRÜCKER & 
NIJSSEN, 1976 AS SPECIES MODEL: A DESCRIPTIVE AND QUANTITATIVE 

APPROACH 
 

 

[Article published] Angulo, A. & Langeani, F. (2017). Gross brain morphology of the 

armoured catfish Rineloricaria heteroptera Isbrücker & Nijssen, 1976 (Siluriformes: 

Loricariidae: Loricariinae): a descriptive and quantitative approach. Journal of 

Morphology, 278, 1689–1705. 

 

 

2.1. ABSTRACT  

 

The gross morphology of the brain of Rineloricaria heteroptera and its relation 

with the sensory and behavioral ecology of the species is herein described and 

discussed. The sexual and ontogenetic intraspecific variation in the whole brain 

length and mass, and within/between the volumes of eight different brain 

subdivisions is also examined and discussed. Negative allometry on the whole brain 

length and mass and on the relative growth of the telencephalon and the optic tecta 

was observed. Positive allometry was observed on the relative growth of the olfactory 

bulbs and the medulla oblongata. Univariate and multivariate statistical analyses did 

not revealed significant differences in the brain subdivisions growth rates among 

sexes and/or developmental stages, except for the optic tectum, and some portions 

of the medulla oblongata; with juveniles and males showing more developed optic 

tecta and medullary subdivisions, respectively. Growth rates for each brain 

subdivision were relatively constant and the slopes of the growth equations were 

almost parallel, except for the olfactory bulbs and the medulla oblongata 

subdivisions, suggesting some degree of tachyauxesis against the entire brain. The 

corpus cerebelli was the more voluminous brain subdivision in most specimens 

(principally adults) and was followed by the optic tectum (the more voluminous 

subdivision in juveniles), the hypothalamus and the telencephalon, in that order. 



	

	

21	

	

Differences in the number of lamellae and in the relative size of the olfactory organ 

also were detected among developmental stages, being more numerous and larger, 

respectively, in adults. Based on these results, it is possible to infer an ontogenetic 

shift in the habitat/resources use and behaviors in R. heteroptera; were the vision, 

primarily routed through the optic tectum, could be fundamental in early stages, 

whereas in adults the olfaction and taste, primarily routed through the olfactory bulbs 

and the medulla oblongata, would play a more important role.  

 

2.2. KEYWORDS 

 

Brain organization, Ontogeny, Sexual dimorphism, Neotropical fish, Teleostei. 



	

	

22	

	

2.3. INTRODUCTION 
 

Brain morphology in teleost fishes is diverse and shows a high degree of 

divergent differentiation hardly comparable to that of other vertebrate groups 

(NORTHCUTT, 1988; KOTRSCHAL et al., 1998; WAGNER, 2003). Along their 

evolutionary history, teleost fishes have been subjected to different selective 

pressures resulting in highly specialized sensory and cognitive capacities, abilities 

and behaviors (NORTHCUTT, 1988; KOTRSCHAL et al., 1998; WAGNER, 2003; 

WHITE & BROWN, 2015); such specializations may be directly reflected, by the way 

of adaptive and exaptive processes, in brain morphology, explaining, in part, the 

huge taxonomic [about 32500 current known species (FRICKEE et al., 2018a)] and 

ecological diversity within the group (PARHAR et al., 2001; WAGNER, 2003).  

To practical purposes, teleost fish brains can be divided into seven main 

regions, areas or structures [i.e., olfactory bulbs, telencephalon, diencephalon, 

midbrain tectum (often referred as the optic lobe or optic tectum), midbrain tegument, 

cerebellum (including the corpus cerebelli and the valvula cerebelli) and medulla 

oblongata (including the electrosensory, facial and vagal lobes); see ALBERT et al., 

1998], each one of which participates in the control of multiple (e.g., in the case of 

multimodal integration centers such as the telencephalon, midbrain tectum, 

cerebellum, etc.) to very specific (e.g., the olfactory bulbs and the electrosensory, 

facial and vagal lobes of the medulla oblongata) sensory and cognitive functions 

and/or behaviors (WAGNER, 2002; BUTLER & HODOS, 2005; ULLMANN et al., 

2010; WHITE & BROWN, 2015). Most of these brain subdivisions show notable 

morphological and functional similarities with the respective brain subdivisions of 

other vertebrate groups (see NORTHCUTT, 2002; WULLIMANN & MUELLER, 2004; 

BROGLIO et al., 2011; WHITE & BROWN, 2015); in other cases, however, 

homologies are still largely unresolved (BUTLER & HODOS, 2005).  

Earlier investigations on fish brain morphology, functioning and development 

(e.g., LISSNER, 1923; GEIGER, 1956) demonstrated that the degree of 

differentiation of the different sensory and cognitive capacities, abilities and 

behaviors in teleost fishes can be correlated, in many cases, with the morphology of 

a certain brain subdivision or area, even more than in other vertebrate groups (ITO et 

al., 2007). As a consequence, brain morphology, in general terms, has been widely 

utilized, in several teleost fish taxa, as a strong predictor for a species (or species 



	

	

23	

	

group) sensory, cognitive and/or behavioral ecology (e.g., BAUCHOT et al., 1989; 

BRANDSTÄTTER & KOTRSCHAL, 1990; KOTRSCHAL & PALZENBERGER, 1991; 

HUBER et al., 1997; KOTRSCHAL et al., 1998; WAGNER, 2001, 2002, 2003).  

Some systematists also have attempted to use the brain morphology as a 

source of “new” characters to identify and/or corroborate phylogenetic affinities within 

some specific teleost fish taxa, in most cases with relative good success (e.g., 

ALBERT et al., 1998; EASTMAN & LANNOO, 2003A, 2003B, 2004, 2007, 2008, 

2011; KHONSARI et al., 2009; PUPO, 2011; ROSA, 2015). This is not surprising 

because neuroanatomical characters are known to provide strong phylogenetic 

information (see SVETOVIDOV, 1953; MILLER & EVANS, 1965; FREIHOFER, 1978; 

KHONSARI et al., 2009). Despite of this, the brain morphology is still relatively poorly 

known and under-explored or under-utilized in the vast majority of teleost fish taxa, 

both in comparative studies and/or as source of characters for taxonomic and 

phylogenetic studies (PUPO, 2011; ROSA, 2015; PEREIRA & CASTRO, 2016).  

In addition to the contribution of the ecological and phylogenetic processes to 

brain morphology variation and differentiation in teleost fishes, and vice-versa, as 

briefly mentioned above, some recent studies on brain evolution and development 

(e.g., ITO et al., 2007; NORTHCUTT, 2004, 2008; GONDA et al., 2009; LECCHINI et 

al., 2014; SALAS et al., 2015; LISNEY et al., 2017) also have recognized the 

fundamental role of the ontogenetic processes. This is also not surprising, because it 

is widely known that the brain, as well as virtually all morphological and physiological 

features of the organisms, scales allometrically with body size (BAUCHOT et al., 

1976; MCMAHON & BONNER, 1983). Most of these studies, however, have used 

correlative approaches at the interspecific level, being that similar intraspecific 

studies are usually rare, despite the fact that they could better determine how 

selection and phenotypic plasticity can influence brain morphology (GONDA et al., 

2011; 2013). 

Despite of such advances, and considering the huge taxonomic and ecological 

diversity within the group, little is known on the variation in brain morphology, as well 

as their evolutionary, sensory, cognitive and behavioral implications, in the vast 

majority of teleost fish taxa. Moreover, as noted by ABRAHÃO & SHIBATTA (2015), 

WHITE & BROWN (2015) and PEREIRA & CASTRO (2016), among other authors, 

more integrative, descriptive and quantitative, approaches, both at the inter- and 

intraspecific levels, are necessary to increase and promote the understanding on the 



	

	

24	

	

diversity, phylogenetic relationships, ecology, physiology and functioning of this rich 

and complex group of organisms. 

Among catfishes, including 39 families and more than 3720 species 

(FRICKEE et al., 2018a), brain morphology, and its ecologic and phylogenetic 

correlations and implications, are relatively well known in the north American 

Ictaluridae, a relatively small family comprised by about 50 species (FRICKEE et al., 

2018a); see for example the contributions of ATEMA (1971; 1980), FINGER (1976), 

KNUSEN (1976), LUNDBERG (1982), TONG & FINGER (1983), LEE & BULLOCK 

(1984), MORITA & FINGER (1985), KANWAL & CAPRIO (1987), STRIEDTER 

(1990; 1991), MEEK & NIEUWENHUYS (1998), NORTHCUTT et al. (2000), among 

others. Among the neotropical catfishes, sensu stricto, including 14 families and 

about 2155 species [≈58% of the total known diversity within the order (FRICKEE et 

al., 2018a)], descriptive and applied studies on brain morphology are, however, 

absent or relatively scarce; see for example the contributions of TRAJANO (1994), 

PUPO (2011, 2015), ROSA et al., (2014), ABRAHÃO & SHIBATTA (2015), ROSA 

(2015), PEREIRA & CASTRO (2016) and ABRAHÃO et al. (2018).  

With the aim of promoting future studies on the brain morphology of 

neotropical catfishes, as well as its ecological, evolutionary and phylogenetic 

correlations and implications, three objectives were raised in the present study using 

the armoured catfish Rineloricaria heteroptera Isbrücker & Nijssen, 1976 

(Siluriformes: Loricariidae: Loricariinae) as species model. These objectives were: (1) 

to describe the gross morphology of the brain of R. heteroptera, as basis for 

comparative studies on brain morphology of members of the Siluriformes and 

specifically of the Loricariinae and Loricariidae; (2) to explore the sexual and 

ontogenetic (post-larvae) intraspecific variation in brain (and the different brain 

subdivisions) size, mass and volume; and (3) comment on the correlation between 

the brain morphology and the sensory and behavioral ecology of the species. 

 

2.4. MATERIAL AND METHODS 
 

2.4.1. Species of study  
 

Rineloricaria heteroptera (Fig. 1) is a medium sized loricariid, reaching about 

133 mm in standard length (SL), restricted to the upper-medium portion of the 



	

	

25	

	

Amazon River basin, near Manaos (state of Amazonas), Brazil (FERRARIS, 2003; 

FROESE & PAULY, 2017), inhabiting mainly small sandy bottom forest streams 

(RAPP PY-DANIEL & FICHBERG, 2008). As most members of the genus, R. 

heteroptera shows a remarkable external sexual dimorphism as well as benthic 

habits (ISBRÜCKER & NIJSSEN, 1976, 1992; ABELHA et al., 2001; DELARIVA & 

AGOSTINHO, 2001; REIS & CARDOSO 2001; LACERDA, 2007; FICHBERG & 

CHAMON, 2008). The genus Rineloricaria Bleeker, 1862, sensu COVAIN & FISCH-

MULLER (2007), with about 63 valid species (FROESE & PAULY, 2017), is the most 

species rich genus within the Loricariinae and also shows the largest distribution 

within the family (from Costa Rica to Argentina), being found in a diverse array of 

environments (COVAIN & FISCH-MULLER, 2007; FROESE & PAULY, 2017). The 

choice of R. heteroptera as the species subject of study was based mainly in the 

availability of (museum) specimens, from different developmental stages and from 

both sexes, for the inevitable and destructive brain removals. 

 

2.4.2. Material examined 
 

Specimens examined (42 in total, 21 males and 21 females; 35.5–114.4 mm 

SL) are deposited at the fish collection of the Departamento de Zoologia e Botânica 

of the Universidade Estadual Paulista, campus São José Rio Preto, São Paulo, 

Brazil (DZSJRP), under the following catalog numbers: 014427, n=3; 014452, n=3; 

014549, n=3; 014657, n=4; 014722, n=2; 014771, n=10; 014909, n=2; 016730, n=2; 

016982, n=3; 017182, n=6; 017371, n=3; and 017429, n=1 [collection data can be 

accessed through the “Species link” online database site (http://splink.cria.org.br/)]. 

Total length (TL), standard length, head length (HL) and body mass were measured 

in all specimens, which were fixed in ≈10% formalin and preserved in ≈70% ethanol. 

Measurements were taken with the aim of a digital caliper (+/- 0.01 mm) and a digital 

balance (+/- 0.01 g).  

 

2.4.3. Data acquisition  
 

Brains were dissected and removed following the protocol described by 

ABRAHÃO & PUPO (2014) as adapted to Loricariidae by ROSA (2015). 

Nomenclature of brain subdivisions follows MEEK & NIEUWENHUYS (1998) and 



	

	

26	

	

BUTLER & HODOS (2005). Because the brains could not be cut off from the spinal 

cord at comparable positions in each specimen, the posterior limit of the brain was 

defined at the posterior end of the alar portion of the medulla oblongata, recognizable 

in all specimens at both dorsal and lateral views, sensu MEEK & NIEUWENHUYS 

(1998) and PUPO (2011), as a small notch or invagination, representing the 

transition zone between this structure and the medulla spinalis (Fig. 2); all cuts were 

made posterior to this limit. After removal from the neurocranium the brains were 

stored in ≈70% ethanol.  

A digital camera attached to a stereomicroscope was used to capture scaled 

images of the brain topography of all specimens; dorsal (before and after removal 

from the neurocranium), lateral (left side) and ventral views were acquired. All 

photographs were taken while the specimens and brains were submerged in water, 

covered to exactly the same depth (i.e., all specimens and brains had about 1 mm of 

water over their surface), and looking straight down onto the brain; therefore, the 

refractive index should have been about the same for all photos.  

Following ABRAHÃO & SHIBATTA (2015) and WHITE & BROWN (2015) a 

total of twenty-eight brain measurements, including the length, width and height of all 

major brain subdivisions [i.e., olfactory bulb, telencephalon, optic tectum, 

hypothalamus, corpus cerebelli and medulla oblongata (see Table 1)], were taken 

from the scaled images using the image-analysis public domain software 

IMAGEJ1.49 (RASBAND, 1997). Total brain length was measured, in dorsal view, 

from the anterior limit of the telencephalon (excluding the nervus tractus olfactorius) 

to the posterior limit of the alar portion of the medulla oblongata (see above). The 

total number of lamellae in each (left and right) olfactory organ, in each specimen, 

also was recorded. 

Volumes of eight major brain subdivisions [i.e., (1) olfactory bulb, (2) 

telencephalon, (3) optic tectum, (4) hypothalamus, (5) corpus cerebelli and (6–8, see 

below) medulla oblongata lobes; Fig. 2] were estimated using the half ellipsoid 

method (modified from VAN STAADEN et al., 1995; HUBER et al., 1997; WHITE & 

BROWN, 2015). This method assumes that each brain subdivision has the shape of 

an idealized half ellipsoid and involves obtaining linear measurements for the length 

(L), width (W) and height (H), converting them into volumetric measures through the 

following formulae: 1/3*π*L*W*H (see VAN STAADEN et al., 1995; HUBER et al., 

1997). Despite  receiving some criticism  [see for example ULMANN et al. (2010) and 



	

	 27	

Table 1. Body length (SL), mass (We) and brain measurements (28) of examined specimens of R. heteroptera (n=42); absolute 

values and percentages of total brain length, head length or total brain volume, as appropriate. Morphometric measurements (-

L=length; -W=width, -H=height), numbered from 1 to 28, are expressed in mm, volumes (“-V”s) are expressed in mm3 and masses 

(“-We”s) are expressed in mg. Linear measurements refer only to the left lobe or counterpart for those bilateral structures; volumes 

for these bilateral structures were doubled, assuming brain symmetry. Other abreviations: Ce=corpus cerebeli, EG=eminentia 

granularis, Hy=hypothalamus, LobVII=lobus facialis, LobX=lobus vagi, M=mean, MO=medulla oblongata (alar portion), 

OB=olfactory bulb, OT=optic tectum, T-=total brain (e.g. T-L=total brain length), Te=telencephalon, SD=standard deviation. 

 

Measurements 
Absolute values 

Percentages of total brain length, or 
total brain volume, as appropriate 

Percentages of head 
length 

Range M SD Range M SD Range M SD 

SL 
35.48–

114.42 
69.32 18.37 – – – – – – 

We 200–8500 2011.9 1682.8 – – – – – – 

(1) T-L 4.10–8.11 6.04 0.88 – – – 
32.90–

66.91 
43.91 6.9 

(2) T-W 2.26–3.68 3.18 0.37 45.40–59.04 52.86 3.41 
14.94–

33.22 
23.26 4.11 

(3) T-H 1.20–2.77 2 0.34 26.40–36.88 32.87 2.22 
11.25–

17.66 
14.33 1.66 

T-V 3.20–23.74 11.34 4.6 – – – – – – 

Continued on the next page…. 



	

	 28	

Measurements 
Absolute values 

Percentages of total brain length, or 
total brain volume, as appropriate 

Percentages of head 
length 

Range M SD Range M SD Range M SD 

T-We 3.23–23.87 11.39 4.57 – – – – – – 

(4) OB-L 0.31–0.84 0.58 0.13 6.83–11.58 9.6 1.16 3.12–5.51 4.18 0.58 

(5) OB-W 0.21–0.74 0.51 0.12 4.30–10.37 8.45 1.18 2.59–5.17 3.66 0.51 

(6) OB-H 0.18–0.72 0.47 0.11 3.69–8.90 7.66 1.07 2.22–4.14 3.31 0.38 

OB-V 0.03–0.93 0.33 0.19 0.54–4.21 2.63 0.87 – – – 

(7) Te-L 0.99–1.78 1.35 0.19 18.71–27.43 22.34 1.98 
7.11–

15.20 
9.81 1.73 

(8) Te-W 0.65–1.15 0.91 0.12 10.47–19.83 15.08 1.55 
4.45–

10.53 
6.63 1.27 

(9) Te-H 0.57–1.04 0.82 0.09 11.12–17.35 13.64 1.18 4.03–9.73 6.02 1.25 

Te-V 0.89–4.44 2.15 0.7 12.77–22.97 18.46 1.77 – – – 

(10) OT-L 1.16–1.96 1.61 0.2 21.21–31.16 27.06 2.11 
7.69–

18.72 
11.9 2.13 

(11) OT-W 0.68–1.08 0.88 0.07 12.56–17.48 14.82 1.34 
4.39–

11.31 
6.57 1.49 

(12) OT-H 0.72–1.29 1 0.14 12.83–20.02 16.49 1.51 
5.17–

10.55 
7.22 1.19 

OT-V 1.32–5.66 3.06 0.92 22.36–32.96 26.5 2.28 – – – 

Continued on the next page…. 



	

	 29	

Measurements 
Absolute values 

Percentages of total brain length, or 
total brain volume, as appropriate 

Percentages of head 
length 

Range M SD Range M SD Range M SD 

OT-V 1.32–5.66 3.06 0.92 22.36–32.96 26.5 2.28 – – – 

(13) Hy-L 1.11–1.69 1.48 0.15 20.56–30.68 24.81 2.74 
6.85–

16.49 
10.98 2.55 

(14) HydW 0.75–1.41 1.15 0.16 15.62–23.36 19.05 1.83 
5.14–

11.78 
8.34 1.37 

(15) Hy-H 0.41–0.87 0.67 0.12 8.68–14.45 11.06 1.26 3.48–7.01 4.84 0.8 

Hy-V 0.83–3.96 2.46 0.8 16.35–25.06 20.94 2.15 – – – 

(16) Ce-L 0.98–1.93 1.52 0.21 21.03–28.18 25.41 1.83 
7.81–

15.37 
11.15 1.8 

(17) Ce-W 1.44–2.64 2.27 0.27 31.66–44.11 37.98 3.12 
10.64–

23.90 
16.71 3.04 

(18) Ce-H 0.63–1.10 0.86 0.11 12.07–17.23 14.4 1.19 4.44–9.89 6.33 1.19 

Ce-V 1.00–5.79 3.22 1.05 22.24–34.20 27.49 2.49 – – – 

(19) EG-L 0.32–0.60 0.45 0.06 6.38–8.81 7.52 0.63 2.37–4.78 3.3 0.56 

(20) LobVII-L 0.73–1.34 1.06 0.14 14.80–21.36 17.61 1.3 
5.45–

10.72 
7.73 1.27 

(21) LobVII-W 0.40–0.82 0.66 0.12 8.34–13.90 10.96 1.24 3.34–5.88 4.77 0.63 

(22) LobVII-H 0.06–0.27 0.14 0.04 1.32–3.51 2.34 0.47 0.69–1.31 1 0.12 

Continued on the next page…. 



	

	 30	

Measurements 
Absolute values 

Percentages of total brain length, or 
total brain volume, as appropriate 

Percentages of head 
length 

Range M SD Range M SD Range M SD 

LobVII-V 0.04–0.62 0.23 0.12 0.82–2.96 1.84 0.49 – – – 

(23) LobX-L 0.39–0.91 0.64 0.12 8.32–12.84 10.52 0.97 3.61–6.26 4.58 0.59 

(24) LobX-W 0.44–0.91 0.73 0.13 9.17–15.29 12.06 1.36 3.67–6.46 5.25 0.69 

(25) LobX-H 0.05–0.24 0.13 0.04 1.20–3.19 2.12 0.43 0.63–1.20 0.91 0.11 

LobX-V 0.02–0.41 0.14 0.09 0.42–2.00 1.12 0.37 – – – 

(26) MO-L 0.41–1.00 0.65 0.15 8.23–13.29 10.68 1.09 3.90–6.06 4.64 0.51 

(27) MO-W 0.40–0.81 0.66 0.12 8.28–13.77 10.84 1.22 3.23–5.84 4.72 0.64 

(28) MO-H 0.05–0.24 0.13 0.04 1.19–3.13 2.1 0.41 0.62–1.18 0.9 0.1 

MO-V 0.02–0.40 0.13 0.08 0.42–1.86 1.01 0.35 – – – 

 
references], this method is still in use in the literature and its objectivity, efficacy and utility, in comparative studies, has been 

corroborated and demonstrated by previous authors (e.g., WAGNER, 2001a, b; LISNEY et al., 2007; GONZALEZ-VOYER & KOLM, 

2010; SALAS et al., 2015, and WHITE & BROWN, 2015, among others); given the above, and according to the objectives herein 

proposed, the use of this methodology, in this particular study, can be justified.  

The volume of the olfactory bulb was measured after removal from the nervus tractus olfactorius. The nervus tractus 

olfactorius, considered to be part of the telencephalon (see MEEK & NIEUWENHUYS, 1998; LISNEY et al. 2017), was not included 

in the volume estimate of the telencephalon (neither in the total brain volume estimate) given the notable variation in their length 

throughout growth (see “Results”). The torus lateralis, the hypothalamus and the lobus inferior hypothalami, sensu MEEK & 

NIEUWENHUYS (1998),  BUTLER & HODOS (2005)  and ABRAHÃO & SHIBATTA  (2015; see Fig. 2),  were measured as a unit in  



	

	

31	

	

the calculation of the hypothalamus volume. Medulla oblongata measurements and 

volumes refer to three different sensory subdivisions: (1) the facial lobe (2), the vagal 

lobe and the (3) alar portion, sensu MEEK & NIEUWENHUYS (1998) and PUPO 

(2011) (see Fig. 2). Only left lobes, or counterparts, were measured to calculate the 

olfactory bulb, telencephalon, optic tectum, hypothalamus and medulla oblongata 

subdivisions volumes; for these bilateral structures the values of the volumes were 

doubled, assuming brain symmetry (see SALAS et al., 2015), for analytical and 

comparative purposes.  

Total brain mass (excluding nerves) was determined using an analytical 

balance (+/- 0.0001 g). Before mass measurement, brains were hydrated in ≈20 ml of 

water, room temperature, for about 1 h, in order to avoid variations due to ethanol 

evaporation. Finally, following SALAS et al. (2015), the total brain volume of each 

specimen was calculated from the total brain mass using the estimated density of the 

brain tissue [i.e., 1.036 mg/mm3; see STEPHAN (1960)].  

 

 
 

Figure 1. Rineloricaria heteroptera; Machado River basin, Madeira River drainage, 

Rondonia, Brazil; (A) entire specimen (DZSJRP 16730; 75.25 mm SL; female); 

dorsal (above), lateral (centre) and ventral (below) views; (B, C) ventral detail of head 

of sexually dimorphic male (B; DZSJRP 14771, 80.15 mm SL) and female (C; 

DZSJRP 17429, 81.11 mm SL); note the presence (in the male) of shorted, thickened 



	

	

32	

	

an curved pectoral-fin spines and numerous small odontodes along the sides of the 

head and the pectoral-fin spines. 

 

2.4.2. Data analyses 
 

The complete dataset was divided into three subsets, one containing lamellae 

counts, other containing body and brain lengths and masses and the third containing 

total brain and brain subdivisions volumes estimates, all of them organized by sex 

and developmental stage. The sex of each specimen was determined, and 

corroborated (after observation of external dimorphic characters, in adult specimens), 

by gonad examination after dissection. For comparative purposes, three 

developmental stages were defined as follow: 1 [early juvenile; <60.0 mm SL (i.e., 

35.0–60.0 mm SL)], 2 (late juvenile; 60.1–80.0 mm SL) and 3 [adult; >80.0 mm SL 

(i.e., 80.1–115.0 mm SL)]; body size at maturation (≈80 mm SL) was defined 

according to published data for other congeneric species with similar maximum body 

sizes (≈130 mm SL) and life cycles (see ERIC et al., 1982; BARBIERI, 1994; 

FROESE & PAULY, 2017). A total of seven males and seven females were 

examined for each developmental stage. Log10 transformations (hereafter LOG) 

were made on the last two data sets in order to (1) improve normality prior to analysis 

and to (2) account for an allometric brain size/mass or brain subdivision volume vs. 

body size/mass or total brain volume relationship (BRANDSTÄTTER & 

KOTRSCHAL, 1990; WAGNER, 2003; SALAS et al., 2015).  

First, a paired sample t-test was performed on the first dataset in order to 

compare the lamellae count values in both left and right olfactory organs in each 

specimen regardless of sex and developmental stage. Parallel, a two-way analysis of 

variance (ANOVA) was performed on the same dataset in order to assess whether 

sex, developmental stage and/or the interaction between them (i.e., 

sex*developmental stage) have (or not) a significant effect on the number of lamellae 

on the olfactory organs (calculated as the mean of both left and right olfactory organs 

values). 

Second, Ordinary Least Squares (OLS) linear regressions, were performed on 

the second dataset in order to assess the scaling relationship between brain and 

body length and masses, respectively; this was done using LOG total brain length 

and mass, respectively, as dependent variables and LOG body length and mass, 



	

	

33	

	

respectively, as independent variables (see LISNEY et al., 2017). Parallel, two 

independent two-way analyses of covariance (ANCOVA) were performed on the 

same dataset in order to assess whether sex, developmental stage or the interaction 

between them have (or not) a significant influence on the scaling of the overall brain 

length and mass; this was done, as in the OLSs, using LOG total brain length and 

mass, respectively, as dependent variables, LOG body length and mass, 

respectively, as covariates, and sex, developmental stage and the interaction term 

between them as fixed factors (see LISNEY et al., 2017). 

Finally, independent OLS linear regressions also were performed on the third 

dataset in order to assess the scaling relationships between the volume of each one 

of the eight major brain subdivisions analyzed [i.e., (1) olfactory bulbs, (2) 

telencephalon, (3) optic tecta, (4) hypothalamus, (5) corpus cerebelli, (6) lobus 

facialis (7) lobus vagi and the (8) alar portion of the medulla oblongata] and the total 

brain volume; this was done using LOG volume of each brain subdivision as 

dependent variable and LOG of total brain volume minus the volume of the respective 

brain subdivision as independent variable [this subtraction was done to account for 

the bias that exists when a brain subdivision is scaled against the total brain volume 

or mass (which already includes the subdivision of interest) (see DEACON, 1990; 

IWANIUK et al., 2010; LISNEY et al., 2017)]. Parallel, eight independent two-way 

ANCOVAs were performed on the same dataset in order to assess whether sex, 

developmental stage or the interaction between them have (or not) a significant 

influence on the scaling of each brain subdivision; this was done, as in the OLSs, 

using LOG volume of each brain subdivision, as dependent variable, LOG of the total 

brain volume minus the volume of the respective brain subdivision, as covariate, and 

sex, developmental stage and the interaction term between them as fixed factors 

(see SALAS et al., 2015; LISNEY et al., 2017). 

Additionally, a principal component analysis (PCA) was performed on the third 

dataset in order to (1) examine the clustering of the samples in the multidimensional 

space and to (2) characterize the patterns of brain organization comparing both 

sexes and developmental stages; this was done using the relative volume of each 

brain subdivision [calculated as a fraction of the sum of the volume of all eight brain 

subdivisions measured in each specimen and normalized using an arcsine square 

root transformation; see SALAS et al. (2015)] as correlated variables, and sex and 

developmental stage as grouping factors; this analysis was performed on an 



	

	

34	

	

autocovariance matrix and using the singular value decomposition method for a 

better numerical accuracy (see SALAS et al. 2015). All statistical analyses were 

carried out using the PAST software version 3.11 (HAMMER et al., 2001). 

 

2.5. RESULTS 
 

2.5.1. Gross brain morphology of Rineloricaria heteroptera (Descriptive 
approach) 
 

Brain subdivisions, and the brain as a whole, are described below in the 

antero-posterior axis direction according to the scheme implemented by ABRAHÃO 

& SHIBATTA (2015) and PEREIRA & CASTRO (2016), with modifications. In paired 

structures only the left lobe or left counterpart is described, unless otherwise stated. 

The encephalon is slightly elongate and narrow, slightly wider in its middle portion 

near the mesencephalon (optic tectum) and the diencephalon, and does not present 

apparent external hypertrophy of the sensory and/or motor regions sensu MEEK & 

NIEUWENHUYS (1998) and BUTLER & HODOS (2005). The brain, as a whole, 

occupies the cranial cavity almost entirely and is, with exception of the olfactory bulbs 

and part of the anterior portion of the telencephalon, completely located ventral to the 

parietosupraoccipital bone and the posterior portion of the frontal bones, and dorsal 

to the parasphenoid and the prootic bones, not contacting directly the Weberian 

apparatus.  

 

Telencephalon 

 

The olfactory organ is located dorsal and anterior to the olfactory bulb, with 

which it is interconnected via the nervus olfactorius (I). The olfactory organ is almost 

rounded, or oval, in shape in dorsal view and presents a total of 12–24 feather-like 

lamellae, which are longer and thicker in the posterior portion of the organ (Figs. 2 

and 3). In larger specimens (>80.0 mm SL) the olfactory organ was usually larger, 

with its dorsal surface larger than the dorsal surface of the telencephalon, whereas in 

smaller specimens (<60.0 mm SL) the opposite pattern was observed (Fig. 3). The 

nervus olfactorius is relatively short [usually shorter than the olfactory bulb length, 

mainly in smaller specimens (<60.0 mm SL)] (Figs. 2 and 3). The olfactory bulb is 



	

	

35	

	

positioned at the anterior portion of the brain and is located ventral to the nasal bone, 

near to the articulation of the lateral ethmoid and the vomer. The olfactory bulb is 

almost spherical in shape (Figs. 2 and 3) and can be sessile (in smaller specimens) 

or pedunculated (in larger specimens). The extension of the nervus tractus 

olfactorius, which connect the olfactory bulb with the telencephalon, varies in 

proportion to the total brain length, being that in larger specimens (>80.0 mm SL) its 

length reaches about 75.0% of the total brain length (T-L; Figs. 2 and 3). The nervus 

tractus olfactorius is inserted directly on the ventral surface of the telencephalon. 

The telencephalon, properly, is located anterior to the optic tectum and the 

hypothalamus and posterior to the olfactory bulb, in both lateral and dorsal views; in 

smaller specimens (<60.0 mm SL) the distal portion of the telencephalon is 

positioned dorsal to the hypothalamus. The telencephalon is longitudinally elongated 

and almost oval in shape, in both lateral and dorsal views; its anterior and posterior 

margins are rounded, being the anterior one slightly smaller than the posterior one. In 

lateral view the medial horizontal axis of the telencephalon form an angle of about 

45° with the medial horizontal axis of the hypothalamus (Figs. 2 and 3). The 

telencephalon is divided in two distinct parts: (1) a conspicuous and well-developed 

area dorsale on its dorsal surface, and (2) a small and narrow area ventrale on its 

ventral surface; in the area dorsale, two regions, the lateromesial and the 

posteromesial, are specially prominent and bulging. In most examined specimens, 

the telencephalon is, comparing to the remaining brain subdivisions (e.g., corpus 

cerebelli and optic tectum in that order), smaller and less voluminous (Table 1). 

 

Diencephalon 

 

The hypothalamus is located at the ventral portion of the diencephalon, 

posterior to the telencephalon and the chiasma opticum, mesial to the torus lateralis 

and the lobus inferior hypothalami and anterior to the adenohypophysis (Figs. 2 and 

3). The hypothalamus is longitudinally elongated and almost oval in shape in ventral 

view; its anterior and posterior margins are rounded, being the posterior margin 

slightly smaller than the anterior one. The lobus inferior hypothalami, located at the 

outer margin of the diencephalon, posterior to the torus lateralis and lateral to the 

adenoypophysis, is longitudinally elongated and almost oval in shape (in ventral 

view); its medial horizontal axis forms, in ventral view, an angle of about 40–50° with 



	

	

36	

	

the medial horizontal axis of the hypothalamus (Figs. 2 and 3). The lobus inferior 

hypothalami is, in appearance, more voluminous than the hypothalamus, but less 

voluminous than the telencephalon and the optic tectum. The adenohypophysis is 

located posterior and ventral to the hypothalamus, on its mid axis, and is almost 

spherical in shape (Fig. 2). 

 

 
 

Figure 2. Topographic brain anatomy of R. heteroptera (DZSJPR 014771), 114.42 

mm SL; Machado River basin, Madeira River drainage, Rondonia, Brazil; (above) 

dorsal, (centre) lateral (left side) and (below) ventral views. Abbreviations: 

Ad=adenohypophysis, Ce=corpus cerebelli, Ch=chiasma opticum, EG=eminentia 



	

	

37	

	

granularis, Hy=hypothalamus, LH=lobus inferior hypothalami, LobVII=lobus facialis, 

LobX=lobus vagi, MO=alar portion of the medulla oblongata, MS=medulla spinalis, 

I=nervus olfatorius, II=nervus opticus, III=nervus oculomotorius, IV=nervus 

trochlearis, V=nervus trigeminus, VI=nervus abducens, VII=nervus facialis, 

VIII=nervus octavus or vestibularis, IX=nervus glossopharyngeus, X=nervus vagus, 

La=olfactory lamella, LLA=nervus linea lateralis anterior, LLP=nervus linea lateralis 

posterior, OB=olfactory bulbs, Of=olfactory organ, OT=optic tectum, 

Te=telencephalon, TL=torus lateralis, Tol=nervus tractus olfactorius, and TC=truncus 

cerebri. Pointed lines indicate the anterior and posterior limits of the brain. 

 

Mesencephalon 

 

The chiasma opticum, where the fibers of the nervus opticus (II) cross the 

horizontal midline of the brain, is located ventrally in the mesencephalon, at the 

posterior half of the telencephalon and anterior to the optic tectum (Figs. 2 and 3). 

The optic tectum is located at the dorsal portion of the mesencephalon, anterolateral 

to the corpus cerebelli in dorsal view, about ventral to the corpus cerebelli in lateral 

view, dorsal to the lobus inferior hypothalami and the truncus cerebri in lateral view 

and posterior to the telencephalon in both dorsal and lateral views (Figs. 2 and 3). 

The optic tectum is composed by two bilaterally-bean shaped lobes, which contact 

both the corpus cerebelli (dorsally) and the lobus inferior hypothalami (ventrally). In 

dorsal view the medial horizontal axis of the optic tectum forms and angle of about 

45° with the medial horizontal axis of the corpus cerebelli; in lateral view each lobe 

appears to be longitudinally elongated and almost oval in shape, with its maximum 

width at the medial portion (Figs. 2 and 3). In a 71.4% of the examined specimens 

belonging to the developmental stage 1 (<60.0 mm SL) the optic tectum (considering 

both lobes) was more voluminous than the corpus cerebelli, in medium sized 

specimens (60.0–80.0 mm SL) the optic tectum was more voluminous only in a 

42.8% of the examined specimens, and in all but one specimen (i.e., 92.8%) 

belonging to the developmental stage 3 (>80.0 mm SL) the corpus cerebelli was, on 

the other hand, the more voluminous brain subdivision (also see the “Quantitative 

approach”).  

 

Rhombencephalon 



	

	

38	

	

 

The corpus cerebelli is located at the dorsal portion of the rhombencephalon, 

posterior to the optic tectum and anterior to the lobus facialis in dorsal view; in lateral 

view its anterior portion is located dorsal to the posterior portion of the optic tectum 

and its posterior portion is located dorsal to the eminentia granularis (Figs. 2 and 3). 

The corpus cerebelli is almost pear-like shaped in dorsal view, with its maximum 

width (at the posterior portion) almost twice the minimum width (at the anterior 

portion); in lateral view, its dorsal profile is relatively straight, without undulations; in 

dorsal view, its lateral and posterior margins are undulated (Figs. 2 and 3). The 

eminentia granularis is usually inconspicuous in smaller specimens (<60.0 mm SL) 

but conspicuous in larger specimens (>80.0 mm SL; Figs. 2 and 3); it is located at the 

dorsal portion of the rhombencephalon, posterolaterally to the corpus cerebelli and 

lateral (at the outer side) to the lobus facialis (Figs. 2 and 3). In lateral view, the 

eminentia granularis is vertically elongated and almost oval in shape, with its vertical 

middle axis forming an angle of about 60° with the medial horizontal axis of the optic 

tectum (Figs. 2 and 3). In smaller specimens (<60.0 mm SL) the eminentia granularis 

is located in an anterior position [contrary, in larger specimens (>80.00 mm SL) it is 

located in a posterior position] and ventral to the corpus cerebelli in lateral view (Figs. 

2 and 3).  

The lobus facialis is located at the dorsal portion of the rhombencephalon, 

posterior to the corpus cerebelli, anterior to the lobus vagi, and mesial to the 

eminentia granularis in dorsal view; in lateral view the lobus facialis is located dorsal 

to the truncus cerebri (Figs. 2 and 3). The lobes of the lobus facialis are 

approximately oval in shape, their longitudinal middle axis form an angle of about 

160° and they do not contact one another (Figs. 2 and 3). The lobus vagi, located 

immediately posterior to the lobus facialis, at the dorsal portion of the 

rhombencephalon, is composed of two “quotation mark” shaped lobes; the anterior 

portion of each lobe is conspicuously intumescent and their posterior portions are in 

contact forming a horseshoe-shaped structure (Figs. 2 and 3). The alar portion of the 

medulla oblongata, located immediately posterior to the lobus vagi, is composed of 

two almost oval-shaped structures (in dorsal view) contacting the medulla spinalis at 

the distal margin; the posterior portion of the alar portion of the medulla oblongata is 

slightly smaller than the anterior portion and both are conspicuously intumescent 

(Figs. 2 and 3). 



	

	

39	

	

 
 

Figure 3. Intraspecific variation on the brain of R. heteroptera; Machado River basin, 

Madeira River drainage, Rondonia, Brazil. (A, B, C) DZSJPR 016730, 38.75 mm SL; 

(D) dorsal, (E) lateral (left side) and (F) ventral views. (D, E, F) DZSJPR 014771, 

54.31 mm SL; (D) dorsal, (E) lateral (left side) and (F) ventral views. (G, H, I) 

DZSJPR 014549, 97.29 mm SL; (G) dorsal, (H) lateral (left side) and (I) ventral 

views. Scale bar=2 mm. 

 

The truncus cerebri is ventrally located in the brain, between the posterior 

portion of the mesencephalon and the anterior portion of the medulla spinalis; it is 

comprised by both the mesencephalon and the rhombencephalon (Figs. 2 and 3). 

Most cranial nerves exit/enter their efferent/afferent fibers, respectively, in this area, 

except, as detailed above, for the nervus olfactorius and the nervus opticus. The 

cranial nerves leaving or entering to the truncus cerebri include: the nervus 

oculomotorius (III), the nervus trochlearis (IV), the nervus trigeminus (V), the nervus 

facialis (VII), the nervus octavus or vestibularis (VIII), the nervus glossopharyngeus 

(IX), the nervus vagus (X) and the nervus linea lateralis anterior and posterior (Figs. 

2 and 3); all these nerves are “unspecialized” and do not present apparent external 

hypertrophy (Figs. 2 and 3). The nerves III–V and VII and the nervus linea lateralis 

anterior exit/enter their efferent/afferent fibers at the anterior-ventral portion of the 

rhombencephalon, at approximately the mid-point of the corpus cerebelli; the nervus 

octavus and the nervus linea lateralis posterior enter their afferent fibers at the 



	

	

40	

	

medial-ventral portion of the rhombencephalon; and the nervus glossopharyngeus 

and the nervus vagus exit/enter their efferent/afferent fibers, respectively, at the 

postero-lateral portion of the rhombencephalon, at approximately the mid-point of the 

lobus vagi (Figs. 2 and 3).  

 

2.5.2. Sexual dimorphism and ontogenetic variation in the number of lamellae 
on the olfactory organs and in the total brain size and mass and brain 
subdivisions volumes in Rineloricaria heteroptera (Quantitative approach) 
 

No significant differences in the number of lamellae between both left and right 

olfactory organs were recorded in the specimens examined (t41 =0, p=1). On the 

other hand, significant differences on the average number of lamellae (considering 

both left and right olfactory organs in each specimen) were recorded between 

specimens of different developmental stages (F2, 41=103.60, p <0.05); i.e., smaller 

specimens (<60.0 mm SL) presented lower average lamellae counts (12–15, mean 

13.7, mode 15), whereas larger specimens (>80.0 mm SL) presented greater 

average lamellae counts (22–24, mean 22.8, mode 23). Contrary, sex and the 

interaction between both sex and developmental stage did not show a significant 

effect on the average number of lamellae (F1, 41=0.03, p=0.85 and F1, 41=0.03, 

p=0.97, respectively); i.e., no differences were observed among sexes, both in 

general and within each developmental stage. 

Both total brain length and mass showed a negative allometric scaling against 

body length (SL) and mass, respectively (Figs. 4 and 5). Total brain lengths and 

masses varied from 4.10 mm (in an early juvenile specimen of 35.48 mm SL) to 8.11 

mm (in an adult specimen of 114.42 mm SL) and from 3.23 mg (in an early juvenile 

specimen with a body mass of 200 mg) to 23.87 mg (in an adult specimen with a 

body mass of 5700 mg; see Table 1); i.e., while largest specimens had a body length 

and mass that was nearly 3.3 and 28.5 times greater than that of the smallest 

specimens, over this same range of body lengths and masses the brain length and 

mass increased only by 2.0 and 7.4 times respectively. The OLS linear regression 

equations, ± standard error of mean (SEM), describing these relationships were: 

y=0.481±0.032 * x – 0.101±0.059 (r2=0.84), for the LOG brain length vs. LOG SL 

relationship (Fig. 4), and y=0.441±0.027 * x – 2.045±0.011 (r2=0.87), for the LOG 

brain mass vs. LOG body mass relationship (Fig. 5); in both cases, no differences in 



	

	

41	

	

the relative growth of the brain were observed among developmental stages or 

sexes, both in general and within each developmental stage; i.e., neither of the fixed 

factors, or the interaction between them, was a significant predictor of the brain 

length or mass change along growth (Table 2). 

 

 
 

Figure 4. Regression analysis (OLS) of the LOG brain size (length) on the LOG body 

size (length) in R. heteroptera. Developmental stages are denoted with different color 

symbols (Green=early juveniles, i.e. <60.0 mm SL; Blue=late juveniles, i.e. 60.1–80.0 

mm SL; and Red=adults, i.e. >80.0 mm SL); females are represented by closed 

symbols, males are represented by open symbols. 

 

As previously stated, OLS linear regressions also were used to assess the 

scaling relationships for each brain subdivision (i.e., olfactory bulbs, telencephalon, 

optic tecta, hypothalamus, corpus cerebelli, facial lobe, vagal lobe and alar portion of 

the medulla oblongata), using the corrected total brain volume as independent 

variable; the resultant regression equations and lines for these relationships are 

presented and illustrated in the Table 3 and the Figure 6, respectively. The 95% 

bootstrapped confidence intervals for the allometric coefficient values for each 

regression equation (see Table 3)  revealed that the slopes of the regression lines for  



	

	 42	

Table 2. Results of the ANCOVAs evaluating the effect of the body length (SL) and the body mass (BW), as the covariates, and the 

sex, the developmental stage and the interaction between them (i.e. sex*developmental stage), as the factors, on the scaling of the 

overall brain length (T-L) and brain mass (T-W) in R. heteroptera. Significant p values are denoted with an asterisk (*). 
 

Brain 
variable 

Specimen variable (SL or 
BW) 

Sex Developmental stage 
Sex*Developmental 

stage 

F1,35 p F1,35 p F2,35 p F2,35 p 

T-L 202.877 <0.001* 0.018 0.894 0.432 0.653 0.605 0.552 

T-W 271.817 <0.001* 0.073 0.787 0.709 0.499 2.032 0.146 

 
Table 3. Slopes or "allometric coeffients" (a) and intercepts or "allometric components" (b), with their respectives ± standard error of 

mean (SEM) and 95% bootstrapped confidence intervals (CIV; N=1999), and correlation values (r2), with their respective associated 

"p" value, of the OLS linear regressions lines for the eigth major brain subdivisions volumes (vs. corrected total brain volumes) in R. 

heteroptera. Other abreviations: Ce=corpus cerebeli, Hy=hypothalamus, LobVII=lobus facialis, LobX=lobus vagi, MO=medulla 

oblongata (alar portion), OB=olfactory bulb, OT=optic tectum, Te=telencephalon. 
 

Brain 
subdivision 

a ± SEM CIV b ± SEM CIV r2 p 

OB 1.9620 ± 0.1303 (1.5640, 2.2781) -2.5942 ± 0.1360 (-2.9471, -2.1596) 0.8500 0.0001 

Te 0.8967 ± 0.0489 (0.7991, 0.9955) -0.5483 ± 0.0474 (-0.6455, -0.4558) 0.8935 0.0001 

OT 0.8117 ± 0.0381 (0.7333, 0.8880) -0.2731 ± 0.0352 (-0.3422, -0.1978) 0.9191 0.0001 

Continued on the next page…. 



	

	 43	

Brain 
subdivision 

a ± SEM CIV b ± SEM CIV r2 p 

Hy 0.9926 ± 0.0575 (0.8367, 1.1421) -0.5722 ± 0.0549 (-0.7200, -0.4217) 0.8816 0.0001 

Ce 0.9732 ± 0.0536 (0.8491, 1.1148) -0.3983 ± 0.0492 (-0.5345, -0.2790) 0.8916 0.0001 

LobVII 1.7094 ± 0.0702 (1.5568, 1.8531) -2.4784 ± 0.0735 (-2.6414, -2.3072) 0.9369 0.0001 

LobX 1.8532 ± 0.0880 (1.6613, 2.048) -2.8578 ± 0.0924 (-3.0743, -2.6417) 0.9172 0.0001 

MO 1.8971 ± 0.0837 (1.6996, 2.0604) -2.9492 ± 0.0880 (-3.1273, -2.7341) 0.9632 0.0001 

 
Table 4. Results of the ANCOVAs evaluating the effect of the corrected total brain volume, as the covariate, and the sex, the 

developmental stage and the interaction between them (i.e. sex*developmental stage), as the factors, on the scaling of each brain 

subdivision volume in R. heteroptera. Significant p values are denoted with an asterisk (*). Abreviations: Ce=corpus cerebeli, 

Hy=hypothalamus, LobVII=lobus facialis, LobX=lobus vagi, MO=medulla oblongata (alar portion), OB=olfactory bulb, OT=optic 

tectum, Te=telencephalon. 

 

Brain 
subdivision 

Total brain volume Sex Developmental stage 
Sex*Developmental 

stage 

F1,35 p F1,35 p F2,35 p F2,35 p 

OB 241.589 <0.001* 2.462 0.126 0.266 0.768 2.316 0.113 

Te 304.305 <0.001* 0.185 0.670 0.258 0.774 0.287 0.752 

OT 503.107 <0.001* 0.156 0.695 3.610 0.038* 0.943 0.399 

Hy 355.804 <0.001* 3.622 0.065 2.334 0.112 2.246 0.121 

Continued on the next page…. 



	

	 44	

Brain 
subdivision 

Total brain volume Sex Developmental stage 
Sex*Developmental 

stage 

F1,35 p F1,35 p F2,35 p F2,35 p 

Ce 317.908 <0.001* 1.252 0.271 0.636 0.536 0.564 0.574 

LobVII 583.533 <0.001* 1.334 0.256 0.826 0.446 0.665 0.521 

LobX 580.415 <0.001* 9.573 0.004* 2.269 0.118 1.624 0.212 

MO 661.137 <0.001* 8.115 0.007* 3.759 0.033* 0.436 0.650 

 
Table 5. Results for the first four components (PCs) of the PCA of the relative volume of eight brain subdivisions in R. heteroptera. 

Other abreviations: Ce=corpus cerebeli, Hy=hypothalamus, LobVII=lobus facialis, LobX=lobus vagi, MO=medulla oblongata (alar 

portion), OB=olfactory bulb, OT=optic tectum, Te=telencephalon. 
 

Factor/Brain subdivision PC 1 PC 2 PC 3 PC 4 

Variance explained, proportion (%) 42.49 24.23 18.77 9.04 

Variance explained, cumulative proportion (%) 42.49 66.72 85.48 94.53 

Relative loadings 

OB 0.621 0.187 0.086 -0.077 

Te -0.092 0.412 0.412 -0.633 

OT -0.456 0.047 0.341 0.567 

Hy 0.020 0.385 -0.787 0.096 

Ce 0.059 -0.802 -0.135 -0.276 

Continued on the next page…. 



	

	 45	

Relative loadings     

LobVII 0.387 -0.031 0.142 0.183 

LobX 0.350 0.000 0.156 0.266 

MO 0.347 -0.014 0.156 0.285 

 

Table 6. Results of the PERMANOVA test for statistical significance between groups of the Loricariidae, based on the PCoA using 

a total of 88 binary or multistate, non-ordered characters describing its gross brain morphology. Pairwise comparisons; F and 

Bonferroni-corrected p (Bp) values. Abbreviations: TG=Taxonomic group; De=Delturinae; Hp=Hypoptopomatinae; 

Hy=Hypostominae; LF=Loricariinae, Loricariini, Farlowellina; LH=Loricariinae, Harttiini; Li=Lithogeninae; LL=Loricariinae, Loricariini, 

Loricariina; Ne=Neoplecostominae; and Rh=Rhinelepinae. Bold numbers denotes a significant difference between groups. 

 

TG (F\Bp) De Hp Hy LF LH Li LL Ne Rh 

De - 0.014 0.306 0.036 0.346 1.000 0.004 0.083 1.000 

Hp 1.427 - 0.004 0.004 0.004 1.000 0.004 0.004 1.000 

Hy 1.675 1.415 - 0.004 0.004 1.000 0.004 0.004 1.000 

LF 1.598 1.481 1.582 - 0.004 1.000 0.004 0.004 1.000 

LH 1.595 1.338 1.480 1.446 - 1.000 0.004 0.007 1.000 

Li 1.499 1.126 1.363 1.334 1.257 - 1.000 1.000 1.000 

LL 1.514 1.530 1.505 1.522 1.448 1.236 - 0.004 1.000 

Ne 1.471 1.358 1.419 1.480 1.446 1.208 1.498 - 1.000 

Rh 1.379 1.031 1.147 1.161 1.188 1.471 1.101 1.074 - 

 



	

	

46	

	

most of the brain subdivisions (with exception of those from the hypothalamus and 

the corpus cerebelli) were significantly different from zero, indicating allometric 

growth (vs. isometric growth). The slopes of the regression lines for the olfactory 

bulbs and the medulla oblongata subdivisions were both >1, indicating that these 

brain subdivisions grow more quickly than the rest of the brain (i.e., they show a 

positive allometry), and the slopes of the regression lines for the telencephalon and 

the optic tecta were both <1, indicating negative allometry. Moreover, comparisons of 

the confidence intervals for the allometric coefficient values of the different regression 

equations (see Table 3) revealed that the slopes for the olfactory bulbs and the 

medulla oblongata subdivisions were not significantly different from each other, but 

they were significantly different from the slopes for all other brain subdivisions, 

indicating some degree of tachyauxesis of these subdivisions against the entire 

brain. The results of the eight independent two-way ANCOVAs, performed on the 

same dataset, revealed that, as expected, the (corrected) total brain volume was the 

most significant predictor of the volume of each one of the eight brain subdivisions 

analyzed (Table 4). In addition, the developmental stage and sex, but not the 

interaction between them, proved to be significant predictors of the volume of the 

optic tectum and the alar portion of the medulla oblongata and of the vagal lobe and 

the alar portion of the medulla oblongata, respectively (see Table 4). As also 

mentioned and illustrated below, smaller specimens (<60.0 mm SL) showed 

proportionally greater (more voluminous) optic tecta and proportionally smaller (less 

voluminous) medulla oblongatas than adults (>80.0 mm SL), whereas, males showed 

proportionally larger (more voluminous) vagal lobes and alar portions of the medulla 

oblongata than females. 

The PCA of the relative volume of the eight brain subdivisions separated 

relatively well in the multidimensional space the three age groups previously defined; 

contrary, no apparent differences in the clustering of the samples were recorded 

considering boths sexes, both in general and within each developmental stage (Fig. 

7). The relative importance of the first four components and the relative loadings of 

each variable in each component are showed in Table 5. The first component (PC1), 

explaining 42.5% of the overall variance, was associated with high loadings for the 

olfactory bulbs volume, and, secondarily, for the medulla oblongata subdivisions 

volumes (i.e., the facial and vagal lobes and the alar portion), and separated 

relatively well adults (>80.0 mm SL), which presents relatively large olfactory bulbs 



	

	

47	

	

and medullary systems, from early juveniles (<60.0 mm SL), which presents, in 

contrast, relatively larger optic tecta (showing negative values in PC1; see Table 5). 

The second component (PC2), explaining 24.2% of the overall variance, was 

associated with positive loadings for the telencephalon and the hypothalamus 

volumes and with negative values for the corpus cerebelli volume; however, the PC2 

did not separate juveniles from adults.  

 

 
 

Figure 5. Regression analysis (OLS) of the LOG brain mass on the LOG body mass 

in R. heteroptera. Developmental stages are denoted with different color symbols 

(Green=early juveniles, i.e. <60.0 mm SL; Blue=late juveniles, i.e. 60.1–80.0 mm SL; 

and Red=adults, i.e. >80.0 mm SL); females are represented by closed symbols, 

males are represented by open symbols. 

 

2.6. DISCUSSION 
 

2.6.1. Sexual dimorphism in the brain of Rineloricaria heteroptera and its 
ecological implications 
 



	

	

48	

	

Most species of Rineloricaria, including R. heteroptera, show a remarkable 

external sexual dimorphism, characterized by the presence (in males) of thickened 

and curved pectoral-fin spines as well as of numerous odontodes along the sides of 

the head, the pectoral-fin spines and the predorsal area (ISBRÜCKER & NIJSSEN, 

1976, 1992; RAPP PY-DANIEL & COX-FERNANDES, 2005). These anatomical 

modifications are usually utilized in agonistic behaviors including intraspecific 

displays and combats (ISBRÜCKER & NIJSSEN, 1976, 1992), and, as expected, 

should reflect in brain morphology leading to sexual dimorphism (PARHAR et al., 

2001). In contrast to the expectated, no sexual dimorphism was found in the relative 

volume of the most of the examined brain subdivisions along development or 

independently of this (Figs. 6 and 7). Significant differences in the average volume of 

the brain subdivisions among sexes, independently of the developmental stage, were 

detected only for the vagal lobe and for the alar portion of the medulla oblongata (see 

above). Interestingly, other studies examining the gross brain morphology of some 

species of neotropical catfishes (e.g., PUPO, 2011; ABRAHÃO & SHIBATTA, 2015; 

ROSA, 2015), cyprinids (Cyprinidae, Cypriniformes; e.g., EVANS, 1931), and cichlids 

(Cichlidae, Perciformes; e.g., VAN STAADEN et al., 1995), which are known to 

exhibit a marked external sexual dimorphism, also do not report sex linked 

differences in most of the examined brain subdivisions along development. These 

results could suggest, as noted by previous authors (e.g., VAN STAADEN ET AL., 

1995; PARHAR et al., 2001; GONZALEZ-VOYER et al., 2009), that sexual external 

dimorphic characters, and related behaviors, have none or little direct impact on the 

external brain morphology of these fishes (at least for the great majority of brain 

subdivisions). On the other hand, as also suggested by previous authors (e.g., 

LANDE, 1980; VAN STAADEN et al., 1995; GONZALEZ-VOYER et al., 2009), 

environmental and ecological factors, which are expected to impose similar selective 

pressures on both sexes, at each different developmental stage, could be more 

important in shaping brain morphology, explaining in part the absence of sexual 

dimorphism in most brain subdivisions.  

MARTINS & LANGEANI (2011) studied the brain morphology of members of 

the Hypoptopomatinae (Loricariidae) and reported wider nostrils in males of some 

genera and species within the subfamily; however, these authors did not mention any 

difference in the olfactory organs, and/or other related brain subdivision, nor discuss 

its possible(s) ethological or ecological implications. MARTINS et al. (2014) and 



	

	

49	

	

ROSA (2015) reported some additional dimorphic neuroanatomical characters in 

members of the subfamily (i.e., larger olfactory organs and a greater number of 

olfactory lamellae in males of some species), but, again, they do not refer to other 

brain subdivision nor discuss its possible(s) ethological or ecological implications. 

The anatomical differences reported in members of the Hypoptopomatinae by 

previous authors, which are possibly related to mate location (see GIBBS, 1991; 

ARIF, 2011), were not observed in R. heteroptera. However, as noted above, 

differences in the average volume of the vagal lobe and of the alar portion of the 

medulla oblongata were observed among sexes. 

 

 
 

Figure 6. Allometric scaling relationships (volume vs. volume OLS regression lines) 

for the eigth major brain subdivisions [Ce=corpus cerebeli, Hy=hypothalamus, 

LobVII=lobus facialis, LobX=lobus vagi, MO=medulla oblongata (alar portion), 

OB=olfactory bulb, OT=optic tectum, Te=telencephalon] in R. heteroptera. 

Developmental stages are denoted with different color symbols (Green=early 

juveniles, i.e. <60.0 mm SL; Blue=late juveniles, i.e. 60.1–80.0 mm SL; and 

Red=adults, i.e. >80.0 mm SL); females are represented by closed symbols, males 

are represented by open symbols. 

 



	

	

50	

	

Medulla oblongata subdivisions are usually associated to the gustative 

perception, receiving sensory information from taste buds present in the barbels, 

mouth, pharynx and gill arches (MEEK & NIEUWENHUYS, 1998; ABRAHÃO & 

SHIBATTA, 2015). Differences in the average volume of these medullary 

subdivisions among sexes can be associated to possible differences in the 

perception and use of food resources and foraging strategies at the intraspecific level 

(EVANS, 1931; ATEMA, 1971). In addition, because several species of Loricariidae 

shows a remarkable “bucal” paternal care, including R. heteroptera (see TAYLOR, 

1983; GROSS & SARGENT, 1985; MARCUCCI ET AL., 2005; RAPP PY-DANIEL & 

COX-FERNANDES, 2005), brain modifications on medullary subsystems such as the 

lobus vagi and the alar portion, also could be associated with increased sensitive 

capacities and/or specialized behaviors related to the care of the progeny [e.g. the 

males of some species of Loricaria and Loricariichthys (not being the specific case of 

R. heteroptera) exhibit an accentuated sexual dimorphism during the reproductive 

season characterized by an increase on the size and structural complexity of some 

oral structures, such as the lips (see TAYLOR, 1983; RAPP PY-DANIEL & COX-

FERNANDES, 2005); additionally, members of these two genera, have massive 

vagal lobes, generally much more developed than in other members of Loricariidae 

(Rosa, 2015)]. However, it is difficult to do more than speculate on the factors behind 

the apparent differences in the volume of some brain subdivisions among sexes 

considering the absence of both detailed life history data for the species under study 

as well as information on the role of the senses in both immature and mature 

specimens, both males and females.  

 

2.6.2. Ontogenetic variation in the brain of Rineloricaria heteroptera and its 
ecological implications 
 

Teleost fish brains grow in a negative allometric relationship to the body 

(GEIGER 1956; RIDET et al., 1977; BRANDSTÄTTER & KOTRSCHAL, 1990; VAN 

STAADEN et al., 1995), and, according to the results of this study (see Figs. 4 and 

5), R. heteroptera is not an exception to this pattern. Brain growth patterns similar to 

that described here for R. heteroptera, i.e., total brain length/mass as dependent 

variable of total body length/mass, have also been documented in some members of 

the Cyprinidae (GEIGER, 1956; BRANDSTÄTTER & KOTRSCHAL, 1990; 



	

	

51	

	

KOTRSCHAL et al., 1990; 1998; KOTRSCHAL & PALZENBERGER, 1991), 

Macrouridae (WAGNER, 2003) and even in elasmobranch fishes (LISNEY et al., 

2007). Additionally, in most of these published cases, as also found for R. 

heteroptera, the measurements related to the brain as a whole (i.e., total brain 

length, total brain width and total brain height) as well as those related to the corpus 

cerebelli, optic tectum and the medulla oblongata, were usually the most variables. 

Moreover, the regression lines reported demostrated that the brain of R. heteroptera, 

as also described for most teleost and non-teleost fish taxa (see KOTRSCHAL et al., 

1990; 1998; WAGNER, 2003; LISNEY et al., 2007; 2017; among others), displays an 

indeterminate growth. 

Specific growth rates for each one of the brain subdivisions in R. heteroptera, 

expressed as volume vs. volume relationships, were relatively constant and the 

slopes of the growth