PISCES. 177 
3. Althougli the mimher of species is small, the diversity of forms is com- 
paratively great. 
4. The genera are identical with those of South America, with the excep- 
tion of ChaldnoimSy which is peculiar to the isthmus [and neighbouring 
parts]. 
6. All the species are carnivorous ; herbivorous fishes are entirely absent. 
[This is not correct ; Pcecilia (^XiphopKot'Ufi) gillii (from the author’s own 
collection) is an herbivorous or at least mud-eating fish j and more such will 
be discovered by further researches,] 
8. Most of the species have not been found elsewhere. 
9. The number of individuals is small. 
10. The greater number of the species are found on the Atlantic and Pacific 
sides of the watershed, viz. : Acanthias vulgaris, Agonostoma elongatnm, mon- 
ticola et nasutum, Macrodon ihicrolepis, Acara cceruleopunctata, Heros sie- 
holdii, Pcecilia gillii, Chalcinopsis striatidus, Tetragonopterus ccneus et gronovii, 
Pimelodus [cinerascem?^, Loricaria uracantha et lima, Hypostomus pie-- 
costomus. 
The author justlj’^ regards the last point as tlie most interest- 
ing result of his researches, but he is mistaken in believing that 
no instance has been known, before Prof. Kner’s examination of 
his collection, of the occurrence of one and the same species on 
both sides of the isthmus (p. 76). Such instances have been 
repeatedly mentioned by the Recorder and Mr. Gill ; and the 
fact has been most distinctly stated in a paper on Central 
American fishes, published in 1861 in Proc. Zool. Soc. Loud. 
Nov. 26. Prof. Wagner explains the fact by the great con- 
striction of the isthmus, the depression of its surface, the pre- 
valent north-eastern monsoon, the flood tide penetrating high 
up the rivers, and the daily passage of aquatic birds from one 
shore to the other. He says that the altitude of the hills be- 
tween 79^^ 29' and 79° 51' long. W. is 206 metres, and that of 
the passes 139 metres only, and that the width of the water- 
shed between the Rio Obispo (a tributary of the Rio Chagres) 
and the Rio Grande is but one-third of a geographical mile. 
In the second part the author treats of the hydrographical 
peculiarities of Ecuador and of the characteristics of the fish- 
fauna of the western slope of the Andes *. He says that 
the alpine forms [Brontes prenadilla) reach here to an altitude 
of 13,400 feet. Arges cyclopum and Brontes prenadilla descend 
to an altitude of 7000 feet ; Trichomycterus laticeps et taenia, 
Pseudo chalceus lineatus, and Saccodon are limited to the region 
between 1000 and 7000 feet of altitude. 
D. Anatomical Publications. 
! Gegenbaur, C. Untersuchungen zur vergleiehenden Anato- 
I mie der Wirbelthiere. Heft 2. Schultergurtel der Wirbcl- 
1 * The author is evidently not completely acquainted with this part of the 
fish-fauna. 
1865. [voL. II.] 
N 
