﻿NORTH AMERICAN NOTOSTRACA — LINDER 15 



with 28 rings, 14 out of 17 having this number. Of course, many more 

 specimens must be examined to get an idea of the variation in this 

 species. 



Of A. cancnfoj-mis, only samples with few specimens have been 

 accessible. One of these is the following lot of 33 females from the 

 vicinity of Skara, Sweden (Uppsala Museum). (No males were 

 found in the sample): 



Number of specimens 



Series of body-rings 



1-- 

 1.. 

 26. 

 1-. 

 2.. 

 1.. 

 l._ 



Of this species, also, no males have been found in Sweden, so it is fair 

 to assume that the population is parthenogenetic. 



The fii'st specimen of A. cancriformis listed, with only 10 thoracic 

 rings, is abnormal, the ninth, tenth, and eleventh pairs of legs coming 

 from the ninth and tenth body rings, an interesting deviation, as it 

 shows that the anterior boundary of the polypody sometimes can be 

 displaced farther foi-ward than the eleventh ring (it is a common 

 occurrence to see it displaced backward). But, in the present case, 

 I think we had better leave this aberrant specimen out of considera- 

 tion. The specimen with spiral growth may be counted as having 33 

 rings, because the two rounds of a spiral probably represent two rings. 

 Thus we have 28 specimens with 33 rings, 2 with 33 -hi, 1 with 34, and 

 1 with 34 -fi rings, which means a quite considerable predominance of 

 specimens with 33 rings. In several smaller samples this tendency is 

 still more pronounced; one lot from the same locality comprised 17 

 specimens, all with 11 + 17+5=33 body-rings. 



Now, in parthenogenetic populations we naturally expect to find 

 a smaller variation than in bisexual ones, if hereditary characters are 

 considered. Thus we can conclude that the number of rings is 

 predominantly due to hereditary influences. 



This conclusion will be strengthened when we consider the environ- 

 mental conditions of Rosenberg's find of populations with a nearly 

 fixed number of rings in the rice districts of California. These con- 

 ditions are probably similar to those of the localities of the same 

 species in Texas and Oklahoma. But in the latter localities we have a 

 wide range of variation in the number of rings, and both sexes are 

 represented in about equal numbers. 



