THE MYRTAPODA OF NORTH AMERICA. 155 



Subfamily III. — Androg-nathinje.* 



Analysis of the genera of Andrognathinw. 



a. Fifth segment normal; pores not raised on a pedicel Dolistenus Fanzago. 



<ui. Lateral carime of fifth segment deeply emarginate; the repugnatorial pores 

 raised on a pedicel, the first placed on the anterior lobe of fifth segment, rest 

 on posterior border of other segments Anduognatiius Cope. 



Subfamily IV. — Platydesminje. 



Analysis of the genera of Platydesmiiice. 



a. Dorsal plates with two large median tubercles, anterior with smaller ones. 



PSEUDODESMUS Pocock. 

 aw. Dorsal plates with two transverse rows of small smooth tubercles. 



PLATYDESMUSt Lucas. 



Order II.-HELMINTHOMORPHA. 



This order is divisible into the following - suborders: 



a. Cardo of mandibles and promentum of gnathochilarium present; segments 30 

 or more Iuluidka.— I. 



art. Cardo of mandibles absent; promentum of gnathochilarium absent ; segments 20 

 or '21; copulation foot of male formed from the anterior pair of seventh seg- 

 ment Polydesmoiuea. — II. 



Suborder I.— IULUIDEA. 



This suborder is divisible into the two following- superfamilies: 



a. Copulation foot of male formed from both pairs of legs of seventh segment ; cardo 

 of gnathochilarium single Juloid.e. — I. 



art. Copulation foot of male formed from the anterior pair of legs of seventh segment ; 

 cardo of gnathochilarium double CAiLiPODOlDiE.J; — II. 



Superfamily I. — Juloidae. 



This contains the following families: 



«. Repugnatorial pores present; anal segment rounded or produced into a spine 

 (ovaries surrounded by a common membrane) Julid.e. — II. 



art. Repugnatorial pores absent; anal segment produced into two slender papillie 

 or unidentate (ovaries not surrounded by a common membrane). 



Craspedosomid.eaS — III. 



*Andrognathinae (= Family Andrognathidce Cope. Pioe. Atner. Phil. Soc, 182, 1869). 



tPlatydesmus Lucas (=Piestodesmus Lucas, =Brachycybe Wood). 



t After comparing Risso's description of Callipus several times with specimens of 

 Lysiopetaluvi I have failed to see Avhy it should not be considered the same as Lysio- 

 petalum. Since Lysiopetalum is a synonym of Callipus,' the family and superfamily 

 names must be made to agree with Callipus. • 



§ Craspedosomidce ( = in part Craspedosomadce Gray, 1842). As this name is based 

 on the oldest genus and besides is older lhau Koch's German name, Chordeumiden, it 

 must be used in place of the latter. 



