560 



SCIENCE. 



[N. S. Vol. XXII. No. 56G. 



under the same general guise of size and 

 appearance some half a dozen distinct but 

 closely related species of Pescado Blanco. 

 In each case the different species most 

 nearly related seemed to be found together, 

 an exception to the rule otherwise almost 

 universal among animals and plants. Later 

 explorations of Dr. Seth E. Meek in this 

 and other lakes confirmed and magnified 

 this anomaly. The genus Chirostoma is 

 confined to the lakes of the tablelands of 

 Mexico. It includes three groups or sub- 

 genera : Chirostoma proper, green in color 

 and with firm, smooth scales; Lethostole, 

 with smaller scales, rough edged, the body 

 white or translucent, and Eslopsarum, of 

 smaller species, also white, but with still 

 larger scales^ larger than in Chirostoma or 

 Lethostole. The species were fomid to be 

 grouped as follows: 



No data exist for the explanation of this 

 peculiar case of distribution. It is possible 

 that C. humholdtianum of the lakes of Mex- 

 ico represents the ancestral type, that the 



groups Eslopsarum and Lethostole have 

 diverged from it, and that the numerous 

 species of the last-named type in the large 

 lakes, Chapala and Patzcuaro, have been 

 formed by mutations in the sense of the use 

 of the word by de Vries. But it is possible 

 that these species have been formed by iso- 

 lation, and that species thus formed have 

 invaded the territory of other species. The 

 shifting of the shores of these volcanic 

 lakes and of the hydrographic basins to 

 which they belong is among the possible 

 causes to be considered. 



Another curious case of the occurrence 

 in one locality of similar species is found 

 in the genus Eviota of the family of Gobi- 

 idae or gobies. 



Eviota contains very minute fishes of the 

 coral reefs, translucent green in color, 

 plain or blotched with orange and marked 

 with black spots. One species one and a 

 half inches long {Eviota ahax) is found in 

 Japan, and one an inch long {Eviota per- 

 sonata) in the West Indies. The other 

 species are all less than an inch in length, 

 some of them but half an inch, perhaps the 

 smallest of all vertebrate animals. One of 

 these, Eviota epiphanes, is found in crevices 

 in the coral reefs of Hawaii. Another, 

 Eviota miniata, is recorded from Guam. 

 All the remaining species, including those 

 most closely related, are known only from the 

 crevices of coral heads in Samoa. Most of 

 the known specimens, hundreds in all, were 

 obtained by the writer and his associates in 

 Apia and Pago Pago. Our native assist- 

 ants would dive for these coral masses, and 

 on cracking them, the little fishes would be 

 found in their channels and interstices. In 

 Samoa the following species occur: Eviota 

 zonura, E. smaragdus, E. prasites, E. afelei, 

 E. sehreei, E. pruinosa, E. herrei, E. dis- 

 tigma. It would seem as if these species 

 could not have had a geographic origin in 

 the ordinary sense, for they are all grouped 



