J. L. BROOKS 89 



population, so that there is less chance that death by starvation 

 will reduce it more drastically. 



The general reproductive pattern of a population determines 

 its genetic system even as the pattern is genetically controlled. 

 The range of genetic variation among the offspring produced 

 by diploid parthenogenesis (of the type with which we are con- 

 cerned) will very nearly be coextensive with that of the parents. 

 Although mutation and crossing over will probably produce 

 some increased variation, this increase in the amount and kind 

 of variation can be expected to be considerably less than that 

 which would occur after sexual reproduction. (See Banta, 1939, 

 for a summary of work by Banta and co-workers on variation 

 following the two types of reproduction.) 



The relative frequency of uniparental and biparental inheri- 

 tance differs according to the environment which the population 

 inhabits. By and large Daphnia are most successful in two types 

 of environments: either large, permanent bodies of water, or 

 small ponds, which become regularly (or irregularly) uninhabit- 

 able because of the disappearance of the water or the formation of 

 ice. A species is usually better adapted to one type or the 

 other, although some can maintain themselves in both. In many 

 lake-dwelling populations, reproduction is almost exclusively 

 uniparental. Sexual forms occur only at infrequent intervals and 

 then comprise only a very small percentage of the population. 

 The offspring produced sexually probably have, in most cases, 

 little effect upon the genetic composition of the population be- 

 cause of the relative unhatchability of ephippial eggs after long 

 periods of uniparental reproduction. Banta (1939) believes that 

 this reduced viability of sexually produced offspring is due to 

 the deleterious recessive mutations that had accumulated during 

 the periods when the population was reproducing by diploid 

 parthenogenesis. However, it is possible that a single recombinant 

 might be more successful than many of the other clonal genotypes 

 which constitute the population. Such an individual could give 

 rise by parthenogenesis to a clone of individuals with similar 

 constitutions. This clone, if truly superior in that particular 



