110 DIPLOPODA. 
the subdivision of this group into families, subfamilies, and genera, is one about which 
there still exists the widest divergence of opinion amongst systematists. 
The most recent attempt to monograph the entire group is that of Attems (Denk. 
Akad. Wien, lxvii. & Ixviii., 1900-1901). In this valuable and useful work only one 
family, the Polydesmide, is admitted. ‘This is divided into the following subfamilies : 
(1) Strongylosomine, (2) Sulciferine, (3) Leptodesmine, (4) Eupolydesmine, (0) Tra- 
chelodesmine, (6) Eurydesmine, (7) Oxydesmine, (8) Eurytropine, (9) Cryptodesmine, 
10) Pyrgodesmine, (11) Cyrtodesmine, (12) Oniscodesmine, (13) Spheeriodesmine. 
This classification, however, is open to criticism from several standpoints. In the 
first place, many of the names employed to designate the subfamilies are inadmissible. 
For example, Sulciferinee cannot stand because of its derivation from the so-called new 
genus Sulciferus, which came into literature still-born. It was proposed for a number 
of species for which two generic names, viz. Anoplodesmus, Poc., and Prionopeltis, Poc., 
to which subgeneric rank was given, were already in use. It may be regarded as a 
synonym of either one or the other of these, and the group, which is probably valid 
and was dismembered from the Strongylosomatide of Cook, may be called the 
Anoplodesminez or Anoplodesmide according to the fancy of authors. Similarly, the 
names Eupolydesmine and Kurytropine cannot be maintained, because there are no 
such genera as Eupolydesmus and Eurytropis contained in the groups they respectively 
symbolise. Moreover, at least four of the genera referred to the Eurytropine had 
been previously made the types of special families by Cook ; and one of these names, 
at all events, must take the place of “ Eurytropine.” The Eupolydesmine must be 
called Polydesmine or Polydesmide. 
Another criticism that must be offered is that a considerable number of admittedly 
good genera, notably Platyrachus and Fontaria, appear, in Attem’s opinion, to be 
only doubtfully referable to the subfamilies to which they are attached. If special 
subfamilies had been made for their reception, or rather if family or subfamily names 
already proposed for them by Cook had been retained, a more satisfactory and consisteut 
result would have been achieved. And justification for this course is amply forth- 
coming in the monograph under discussion ; for if the differences between, for example, 
the Strongylosomini, Anoplodesmini (Sulciferini), Leptodesmini, Polydesmini (Kupoly- 
desmini), and Trachelodesmini be taken as the criteria for groups of this rank, many 
of the doubtfully placed genera might have been logically accommodated in the same 
way; and results not very different from those obtained by Cook in 1896 would have 
been arrived at (Ann. N. York Acad. Sci. ix. pp. 4, 5). Although in this work Cook 
made no attempt to characterise the families he erected, their names, where based 
upon defined generic forms, must be retained, if the grouping of the genera into families 
er subfamilies be admitted. 
In the present state of our knowledge of the Polydesmoidea it does not appear to me 
