GILBERT AND STAEKS — FISHES OF PANAMA BAY 



89 



There seem to be no differences, save those of color, between A. dovii and A. 

 reti'osella. The fin-rays, general proportions, and squamation seem alike in the two. 

 The fin-formula is: dorsal VI-I, 9; anal II, 8. The lateral line traverses 24 or 

 25 large scales to the base of the caudal fin, and 5 or 6 scales of reduced size on the 

 fin itself. Between the lateral line and the base of the spinous dorsal are 2^ rows of 

 scales (as in retrosella). In A. atricaudus there are 4i or 5-i- rows, in A. atridorsalis 

 3-2- rows between the lateral line and the spinous dorsal. 



Measurements in Hundredths of Length without Caudal. 



Species 



Locality 



Length without caudal in mm 



Head 



Depth 



Eye 



Interorbital (bone) 



Snout  .' 



Maxillary 



Third dorsal spine 



Second dorsal ray 



Second anal spine 



Second anal ray 



Pectoral 



Ventral 



Caudal 



APOGON DOVII 



Panama 



60 

 40 



34 



I2i 



9i 



20i- 



17 



24 



lU 



20 



25 



20i- 



28 



A. RETROSELLA 



Mazat- 

 lan, 

 Mex. 



76 

 40 



34 

 13 

 8^ 



9 

 21 



17 



25 



I2l 

 21 



25 

 22 

 28 



San 



Benito 



Id., Mex. 



55 

 42 



34 

 12 



9 



9 



21 



17 

 24 

 12I 

 21 



24 

 20 



29 



Family CENTEOPOMID.E. 

 160. Centropomus undecimalis (Block). 



Frequently found in the Panama market. 



We are unable to follow Dr. Jordan (1895 h, p. 452) in recognizing the 

 Pacific form as a distinct species (C. viridis). The separation is based on the supposed 

 greater length of the appendages to the air-bladder in viridis, and the shorter second 

 anal spine. We find, however, that the third anal spine projects beyond the second 

 in Atlantic specimens as well as in those from the Pacific, and there seems to be no 



