﻿30 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 102 



however, is highly advisable if one wants to get an accurate result 

 with a minimum of effort. 



When the niunber is actually mentioned in the literature, it appears 

 that only one or two specimens have been counted, and a fixed 

 number is given for each species. I have foimd only one exception 

 to this: Chamberlin and Duncan (1924) noticed a variation in the 

 number of leg-bearing rings (p. 9) and also mentioned that the num- 

 ber of pairs of legs in L. ardicus is 41-46. 



In fact, the number is not constant in any of the species of which 

 I have examined a reasonable number of specimens, though it seems 

 to be fairly constant in parthenogenetic populations. The American 

 species present the following figures: L. ardicus 41-46, L. packardi (1 

 specimen) about 35, L. couesii 35-40, L. lemmoni unknown, L. bilo- 

 batus (1 specimen) about 62, the new species of Lepidurus 60-71, 

 the new variety of Lepidurus 67-68 (2 specimens), and A. longicaudatus 

 54-66. In two of the above species approximate figures are given 

 because the specimens in question were not in a sufficiently good 

 state of preservation to allow an accurate counting of the legs without 

 causing some damage to the specimens, and, the specimens being 

 imique, dissection did not seem advisable. In some species, at 

 least, the females usually have a few more legs than the males. 



We notice that the greatest number of legs is found not in Apus, 

 where the leg-bearing abdominal rings may be as many as 21, but 

 in a species of Lejyidurus, where these rings do not number more 

 than 18. Within the latter genus, however, all species with 14.5-18 

 leg-bearing rings have 60-71 pairs of legs, and the species with 9.5-13 

 leg-bearing rings have only 35-46 pairs. For counts of extra- American 

 species, I refer readers to table 3, in which some small changes may be 

 necessary when more specimens are counted. 



It seems clear that the number of legs is a taxonomic character of no 



small importance in many cases, and that we have to take the trouble 



to count them. 



NORTH AMERICAN SPECIES 



Genus LEPIDURUS Leach 



Lepidurus Leach, 1816, p. 259. — Simon, 1886, p. 428 (earlier synonj'my), 

 Bilobus SiDOROV, 1924; 1927. 



The main character of the genus is the presence of a supra-anal 

 plate on the telson, and this seems to be a reasonably good character 

 for practical pm'poses (p. 26). Further, the first pair of legs have 

 comparatively short endites, not much projecting beyond the margin 

 of the carapace (in the new species from Washington, however, they 

 are quite long), the legless rings are 3-8, mostly 4-6 in number, and 

 the ova are reported as larger than in Apus. 



The second maxiUae are weU developed, not reduced in size or absent 

 as m most species of Apus. 



