136 



ONTOGENY AND SYSTEMATICS OF FISHES-AHLSTROM SYMPOSIUM 



0.9 



E 

 E 



c 



O) 



c 



(V 



-J 



0) 



o 



c 



3 

 T3 



(U 

 Q. 



CI3 Cypriniformes 

 I 1 Characiformes 

 fTTTTI Siluroldel 

 — - Gymnotoidel 



— I 1 1 1 1 1 1 — 



0.0 0.2 0.4 0.6 



Size (PCI score) 



Fig. 69. Morphometric characters important in defining major groups of ostariophysan larvae. Shaded areas and individual Imes enclose all 

 regression-predicted values at two standard sizes (0.0 and 0.6) of a given taxon. 



included: (11) Size at hatching (PCI score at total length for 

 hatching, based on the regression of PCI on the logarithm of 

 total length), (12) Size at complete finfold absorption (PCI score 

 at total length for complete finfold absorption, based on the 

 regression of PCI on the logarithm of total length), (13) Yolk- 

 sac shape (ratio of the greatest vertical length [depth] of the yolk 

 sac to its greatest horizontal length in recently hatched individ- 

 uals). 



Meristic characters.— These include: ( 1 4) Preanal myomeres (all 

 myomeres at least partly anterior to a vertical line projected 

 from the anus, including an occipital segment) and (15) Postanal 

 myomeres (all myomeres entirely posterior to a vertical line 

 projected from the anus, including a urostylar segment). 



Missing myomere data for Hoplias malabancus were taken 

 from vertebral counts made from radiographs of adults (UMMZ 

 66435). The one-to-one ontogenetic relationship of myosepta 

 to neural spines in monospondylous fishes (Lauder, 1980) per- 

 mitted estimation of myomere number from vertebral number 

 only by inclusion of myomeres for an occipital segment, a uro- 

 stylar segment, and the four (five in siluroids) obscured We- 

 berian vertebrae (Fuiman, 1982a). 



Ontogenetic characters. —Size, rather than chronological age, is 

 most closely related to development (Gerking and Rausch, 1 979). 

 Thus, total length at the onset of selected developmental events 

 was recorded. To compare these sizes among species with dif- 

 fering initial lengths and ranges of lengths for the larval period. 



the logarithm of the hatching length was subtracted from the 

 logarithm of the length at a given event. This difference was 

 divided by the difference of the logarithms of length at complete 

 finfold absorption and at hatching (the criteria used here to 

 delimit the larval period). The resultant character was the per- 

 centage of the larval period that occurred prior to the event, an 

 estimate of relative timing of the event. When characters were 

 present at hatching or did not develop until after complete fin- 

 fold absorption they were coded as 0.00 or 1.10, respectively. 

 The following events were recorded: (16) Anal fin rays (first 

 distinct ray), ( 1 7) All median fin rays (all median fin rays present, 

 finfolds may persist, fin margins may be incomplete), (18) Yolk 

 absorption (complete absorption of yolk), (19) Head straight 

 (head free from yolk sac and not deflected downward), (20) Eye 

 pigment (first uniform pigmentation of retina). 



Presence/absence characters— Presence (coded as I ) or absence 

 (0) of the following structures at any time during the larval 

 period was recorded: (21) Jaw teeth (teeth on the premaxilla, 

 maxilla, or dentary), (22) Adipose fin, (23) Caudal spot (con- 

 gregation of melanophores at the base of the caudal fin forming 

 a distinct spot), (24) Lateral stripe (melanophores on the mid- 

 lateral myoseptum forming a continuous, longitudinal stripe). 



Phylogenetic results 

 The Wagner tree (Fig. 67, Table 28) contains 101 steps for 

 the 46 two-state factors ("characters"). Members of each major 



Table 29. Distribution Statistics of Preanal. Postanal, and Total Myomere Number for Ostariophysan Larvae. Values are based on 



means for each species. 



Preanal myomeres 



Postanal myomeres 



' Including Cyprinidae, Calostomidae, and Cobiloidea. 



Total myomeres 



