Section 9 — Population Genetics 



9.53. On the Influence of the Number of Inbieeding 

 Generations on Viability and Size in Droso- 

 phila melanogaster. F. A. Lints (Louvain, 

 Belgium). 



As pointed out by Lerner (1954) attempts to 

 verify the general idea that heterozygotes are 

 better canalized than homozygotes are urgently 

 needed. A wild strain was submitted to in- 

 breeding immediately after capture, and in 

 order to obtain also reciprocal hybrids, each 

 generation was backcrossed to the wild strain 

 Swedish-B. Twelve generations were realized, 

 seven of which were measured for several 

 traits at five temperatures ranging from 30° to 

 18 C. This study concerns only viability and 

 size. A full report is actually in preparation and 

 will be submitted to Heredity. 



Viability: In absolute values the viability of 

 hybrids does not vary during the successive 

 generations, but in relative terms, and with 

 relation to the control passes from a superior to 

 an inferior level. There is no relation between the 

 number of inbred generations and the percentage 

 heterosis for viability. The variance of inbreds 

 increases irregularly between the first and the 

 12th generation; the effects of inbreeding on 

 viability can be more adequately described by the 

 regression viability-temperature which increases 

 regularly with the number of inbred generations. 



Size: The coefficient of variation of the size 

 for inbreds and hybrids, considered in the 

 complete array of environments used, is identical. 

 This unexpected fact is due, for the hybrids, to 

 a smaller variance in some of the environments 

 considered, combined to a significant regression 

 in function of temperature, for the inbreds, to a 

 large variance at each temperature and to a 

 quasi absence of temperature-size regression. 



Conclusion: The incompatibility of these data 

 with the classical hypotheses of additivity and 

 of heterozygosity per se have to be stressed. In 

 view of the actual and earlier results inbreeding 

 degeneration has to be considered at two levels 

 one at which it has its maximal effect, on viability 

 namely, the other at which it is probably less 

 active, namely on various luxuriance and per- 

 haps on fitness traits of the hatched individuals. 



9.54. Reproductive Isolation between Incipient Species 

 of Drosophila paulistorum in Brazil. Chana 

 Malogolowkin and A. Solima Simmons 

 (New York, U.S.A.). 



Two races or incipient species of the D. paulis- 

 torum complex are known to occur in Brazil, the 

 Amazonian in the states of Amazonas and 



Para and the Andean-Brazilian in the territory to 

 the south. At Belem, Para, both races occur 

 sympatrically. The strains of the two races are 

 difficult to cross, and when crossed produce 

 progenies consisting of fertile daughters and 

 sterile sons. A number of so-called "transitional" 

 strains were however found in the states of 

 Bahia and Ceara; they belong to the Andean- 

 Brazilian race, but they can be crossed and 

 produce fertile hybrids with some strains of the 

 Amazonian race. Sexual (ethological) isolation 

 has been studied, by means of the "male choice" 

 method, between various strains of the two 

 races. Females of the transitional strains show a 

 significantly weaker isolation from Belem 

 Amazonian strains than do females which 

 produce sterile hybrids with the latter. The 

 Amazonian females from Belem do not however 

 accept males of the transitional strains any more 

 easily than do males of the other strains of the 

 Andean-Brazilian race. 



The results as a whole are consistent with the 

 assumption that sexual isolation is built up, or 

 at least strengthened, by natural selection in 

 territories where two races coexist sympatrically. 

 Territories inhabited by only one race have 

 populations which show relatively weaker 

 isolation when exposed to the challenge of 

 hybridization with other races. 



9.55. The Races of Drosophila rubida. W. B. Mather 

 (Brisbane, Australia). 



The species D. rubida Mather of the immigrans 

 species group occurs in northern Queensland, 

 New Guinea, New Britain and New Ireland and 

 is morphologically uniform. However, it has been 

 shown to be rich in inversion polymorphism by 

 mating males and despermed females from the 

 wild to an inversion free standard strain from 

 Northern Queensland and scoring seven larvae 

 per cross against a photographic giant chromo- 

 some map prepared from the standard strain. 

 By these methods eleven simple and three 

 complex inversions have been detected. 



Sexual isolation tests between cultures from 

 Cairns (Northern Queensland), Port Moresby, 

 Madang, Lae and Samarai (Eastern New Guinea) 

 and Rabaul (New Britain) have been carried out 

 by confining ten males of one strain with ten 

 females from another and after a food change at 

 five days dissecting the female genital tract for 

 the presence of sperm after a total of ten days. 

 By these methods it has been shown that where- 

 ever Rabaul males are involved in the cross the 

 sexual isolation is high (usually less than 50 per 

 cent insemination) and that sometimes when 



161 



