130 DIPLOPODA. 
Fam. PYRGODESMIDZ. 
Stylodesmide, Cook, Ann. New York Acad. Sci. ix. p. 5 (1895) ; Proc. U.S. Nat. Mus. xviii. p. 82 
(1895) ; Amer. Nat. xxx. p. 418 (May 1896) (in part.). ; 
Pyrgodesmide, Silvestri, Ann. Mus. Genov. (2) xvi. p. 192 (March 1896) (in part.). 
Stiodesmide, Cook, Brandtia, v. pp. 20 & 25 (June 1896). 
Pyrgodesmine, Attems, Denk. Akad. Wien, Ixviii. p. 375 (1900). 
Decaporodesmide, Kenyon, Proc. Ent. Soc. Wash. iv. p. 299 (1899). 
Anterior border of first tergal plate produced into a semicircular, grooved, and lobate crest completely 
concealing and overlapping the head and forming a continuous series with the keels of the succeeding 
segments ; central part of the plate elevated and much higher than the crest. Head with a densely 
granular frontal area, Antenne clavate, the fifth segment the longest and thickest. Dorsal surface of 
segments elevated, granular, with some larger tubercles forming a pair of longitudinal crests *, one 
on each side of the middle line, frequently a similar but smaller crest between this and the base of 
the keels. Keels rising low on the sides, depressed or nearly horizontal, marginally grooved on the 
dorsal surface, with lateral and posterior borders lobate, the posterior lobe of the lateral border carrying 
the pore, when present, frequently on a papilla. Pores commonly suppressed on the 17th, 18th, and 
19th segments, sometimes on others. Tergal plate of 20th segment moderately broad, not concealed by 
the keels of the 19th. Sterna very narrow. Basal portion of phallopods enlarged to form an arched 
hollow receptacle, in which the distal portion is more or less conccaled. 
Distribution. Tropical parts of America, Africa, and Asia. 
I have adopted for this family the name Pyrgodesmide proposed by Silvestri, because, 
as Attems states, Sty/odesmus appears to be a synonym of Urodesmus of Porat, and a 
family name must be derived from the one in use for one of the genera contained in 
that family. Again, even if in the future it be found that the type of Stylodesmus 
differs from that of Urodesmus in characters to which generic status can be given, it 
is quite possible that the view may be taken that Urodesmus, Stylodesmus, Herco- 
desmus, and others, in which the 19th segment is enlarged so as to conceal partially 
the reduced 20th segment, should constitute a family distinct from those genera, like 
Lophodesmus, Pyrgodesmus, and Psochodesmus, in which the 19th and 20th segments 
are normally constructed. 
That this was the opinion of Cook in 1896, though not in 1895, appears from his 
latest paper on this group published in ‘ Brandtia,’ where the family name Hercodesmide 
is given to the genera with abnormally formed posterior segments and Stiodesmide to 
those in which they present the usual conformation ; and it may be that his substitution 
of the name Hercodesmide for Stylodesmide was due to the discovery that Stylodesmus 
sinks into synonymy under Urodesmus. ‘This, however. is not at all clear from his 
writings. In any case the name Stiodesmide is of Jater origin than Pyrgodesmide ; 
and the latter should, in my opinion, be used for the two genera recorded in this 
monograph, unless Pyrgodesmus with its single median turret-shaped dorsal processes 
* In the Ceylonese genus Pyrgodesmus, Poc., the two crests of tubercles are represented by a high 
upstanding turret-shaped process. 
