ECOLOGY AND ISOLATION 



611 



but critical data are difficult to find. 

 Even habitat isolation partakes of some 

 micro-spatial or micro-topographic separa- 

 tion of populations with little dispersion, so 

 that such isolation is only different in de- 

 gree from the grosser geographical types 

 discussed under other headings. Topogra- 

 phy may separate genetically identical pop- 

 ulations, but ecologic factors usually sepa- 

 rate genetically different populations that 

 have already been guided by natural selec- 

 tion toward adaptive divergence, even 

 though the populations may not be wholly 

 isolated reproductively (i.e., with resultant 

 ecologic subspecies or ecotypes, p. 626). If 

 selection pressure is greater than the dis- 

 persal and cross breeding between partially 

 isolated populations, divergence may occur 

 (p. 616). If dispersal and cross breeding 

 outweigh selection pressure, divergence will 

 not occur. Other types of isolation may 

 augment habitat isolation in any given 

 case. 



Ecologic habitats in close geographic 

 proximity on occasion have closely related 

 species, which have recently diverged in 

 their adjustment to the environment. For 

 instance, the mosquito larvae of Anopheles 

 melas and A. gambiae from the coasts of 

 Africa are limited by different degrees of 

 salinity in the water medium, and such 

 divergence could conceivably result from 

 selection and habitat isolation (Ribbands, 

 1944). A. melas and A. gambiae are both 

 adjusted to fresh water, but only A. melas 

 can tolerate brackish water in which the 

 salinity exceeds 50 per cent of that of sea 

 water. Selection toward adaptation to a 

 brackish water habitat with its consequent 

 isolation from the other species may well 

 have been responsible for ultimate specia- 

 tion. A. melas usually occurs in brackish 

 water under natural conditions in spite of 

 its tolerance of fresh water under experi- 

 mental conditions. When the two species 

 occur together in fresh water, A. gambiae 

 seems to compete more successfully than 

 A. melas. 



Another case of two widely overlapping 

 geographic species exhibiting habitat dif- 

 ferences occurs among the crabs of the 

 genus Ocijpode, found along the Pacific 

 coasts of Central and South America 

 (Crane, 1941a). One species, O. gaudi- 

 chaudii, lives on protected beaches and on 

 the shores of lagoons, feeds on minute or- 



ganic matter in the sand, and is active only 

 in the daytime. The other species, O. occi- 

 dentalis, lives on unprotected beaches 

 beaten by heavy surf, is a predator and 

 scavenger, and is active in the adult stage 

 only at night. Only occasionally are these 

 species found on the same beach, but in 

 each instance of ecologic overlap noted, 

 one or the other is present only in the 

 young stages. It is possible that some form 

 of ecologic isolation was involved in the 

 initial separation of these two species. Both 

 diel and habitat isolation are now well es- 

 tablished, together with adaptive differ- 

 ences in feeding. 



Perhaps one of the most striking cases of 

 habitat isolation and speciation without any 

 considerable degree of topographic isola- 

 tion at the present time is illustrated by 

 the fishes of certain African lakes (Myers, 

 1936; Worthington, 1937, 1940). Since the 

 arid period of the Pleistocene, preceding 

 the last glaciadon in Europe, adaptive ra- 

 diation (pp. 664, 700) of the fresh- water 

 fish genus Haplochromis seems to have oc- 

 curred in Lake Victoria, which is nowhere 

 more than 75 meters deep. This genus has 

 a large series of endemic species ranging 

 from some with short blunt mouths and 

 minute hairlike teeth adjusted to feeding 

 upon small plants and animals, to others 

 with long protrusible mouths, large under- 

 hung jaws, and large teeth adjusted to 

 feeding on other fishes. Some species have 

 large flat-crowned teeth adapted for crush- 

 ing mollusk shells. 



Fifty-eight endemic species of the family 

 CichUdae are found in Lakes Victoria and 

 Kioga— an indication of the amount of adap- 

 tive evolution during the 15,000 to 20,000 

 years or more since these lakes were 

 formed. One hundred and seventy-five spe- 

 cies of cichlid fishes belonging to twenty- 

 three genera (twenty endemic) are re- 

 ported from Lake Nyasa (Trewavas, 

 1935). One hundred and one of these spe- 

 cies belong to the genus Haplochromis. 

 Trewavas says, "the Nyasa species are a 

 natural group and may, perhaps, have 

 evolved in the lake from a single ancestral 

 form." Five species of Haplochromis are 

 found in different ecologic niches in Lake 

 Albert (Trewavas, 1938). Mayr (1942, p. 

 215; 1947) makes an alternative sugges- 

 tion that the large lake could conceivably 



