NO. 1673. • Parasitic coPEPODs— WILSON. 393 



strings are bent inward as soon as they emerge from the genital seg- 

 ment and are brought together on the mid-line. They then turn 

 backward side by side and so close together that they are usually in 

 contact. 



In those species whose anal laminae have well-developed wings on 

 the inner margin the egg-strings pass backward between the two 

 wings. This brings the thickened conical outer margin of the lam- 

 inae outside the egg-strings and thus holds them in together. 



The distinguishing characters of a male Pandarus are the large 

 secondary lobes arising from the posterior border of the carapace 

 inside the regular posterior lobes, the two pairs of rudimentary swim- 

 ming legs on the genital segment, the two-jointed abdomen with the 

 joints of equal length, and the fact that all the rami of the swimming 

 legs are two-jointed. 



Secondary lobes are found on the posterior margm of the carapace 

 in some of the other genera also (Perissopus, Nesippus, etc.), but 

 they are much smaller than m Pandarus, and are easily overlooked, 

 while here they are prominent in all the species examined and one 

 of the first characters that would be noticed. 



In most of the other Nogaus males there are no rudimentary legs 

 visible on the genital segment; here in each of the known species 

 there are two pairs, well defined and promment. 



Their presence is indicative that the so-called genital segment is 

 really a fusion of two segments, the fifth and sLxth thoracic seg- 

 ments, each wdth its pair of legs. This idea has been already advanced 

 by the author,*^ and it receives particular confirmation here, where 

 evidences of fusion are shown also in the genital segment of the 

 female. Scattered testimony was furnished by the two pairs of legs 

 on the genital segment of some Caligus species (for mstance, isonyx, 

 pelamydis, stromatei, etc.) and of many LepeopJitheirus species (for 

 instance, nordmannii, Mppoglossi, edwardsi, d.issiniulatus , etc.) and 

 in the structure of the genital segment in the male of the genus 

 Homoiotes. Here among the Pandarinse the segment itself is plamly 

 differentiated in Dinematura, and is indicated by the rudimentary 

 plate in Pandarus and EcMlirogaleus. 



With this accumulation of evidence we can no longer doubt that 

 there are really six segments in the thorax of all the Caligidae, two of 

 which, the fifth and the sixth, are ordinarily so thorougiih^ fused as 

 to be indistinguishable. Wlien only one pair of legs is visible on the 

 genital segment it is usually the sixth pau' at the posterior corners, 

 instead of the fifth pair, as we have been calling them. 



In 1861 Steenstrup and Liitken suggested that the genus Pandarus 

 ought to be separated into two subdivisions — one to include the true 

 genus Pandarus, made up of Pandarus crancliii as a type, together 



oProc. U. S. Nat. Mus., XXVIII, p. 662. 



