104 UK- G. CHILTON OK THE 



c. Uropoda not projecting beyoud terminal segment ; animals rolling into 



perfect ball Akmadilliid^. 



While this table does not profess to be any very near approach to a natural classifi- 

 cation of the Terrestrial Isopoda, it is probable that the two large divisions A and B do 

 represent distinct groups in which development has proceeded on similar lines, in each 

 case leading from animals living on the sea-shore within reach of the waves and 

 breathing only very moist air, to others of pure terrestrial habits capable of breathing 

 ordinary dry air, i. e. the Helleridae in the one case, and the Armadilliidte in the other. 

 It is interesting to notice, too, how the protection affoi'ded by the animal's power of 

 rolling itself up into a ball has been acquired in different groups that are certainly 

 of independent origin, e. g. in the Tylidae, Hellerida?, Armadillidie, and to a less perfect 

 degree in some of the Scyphacidse {e.g. Acttecia), and perhaps also in some of the 

 Oniscida), and how similar is the general appearance of the body in each of these groups, 

 though of course the detailed arrangements by which it is acquired vary. If we go 

 beyond the Oniscoidea, we find a similar power of rolling into a ball and a somewhat 

 similar external apj)earance in the Sphseromidte and, among the Myriapoda, in the 

 Glonerida3. 



A tabular arrangement of the Oniscoidea, based on much the same characters as 

 I have used, was given many years ago by Ulianin * ; his table, however, goes into 

 greater detail and separates the genera, and imfortnnately it is in the Russian language f. 



For the benefit of those who may wish to identify their specimens without going to 

 the trouble of dissecting out the mouth-parts, I give the following artificial key to the 

 New Zealand genera, and a similar key to the species under each genus represented by 

 more than one species : — 



A. Body convex ; animal capable of rolling into a l)all. 



I. Flagellum of antenna 2-joiuted. 



1. Outer branch of uropoda large and terminal Armadillidium. 



2. Outer branch of uropoda small, inserted on the inner side of the enlarged base. Armadillo. 



II. Flagellum of antenna with more than two joints. 



1. Uropoda quite concealed beneath terminal segment Tylos. 



2. Uropoda extending beyoud terminal segment and visible in dorsal view . . Actteciu. 



B. Body more or less flattened ; animal not capable of rolling into a ball. 



I. Flagellum of autenns many -jointed Lygia. 



II. Flagellum 2-jointed. 



1. Side-plates of metasome large, cxpainled PorcelUo. 



2. Side-plates of nietasome small, adpressed . Metoponort/ms. 



III. Flagellum with three to six joints. 



1. Eyes large, crescent-shaped, of many ocelli Scyphax. 



2. Eyes small, not more than three ocelli. 



a. Body with longitudinal ridges Huplophthalmus. 



* ' Crustacea Turkestriniii',' >St. Petersburg & Moscow, 1S75 (see Budde-Luuil, /. c. p. i'J.). 



t I liiive to thank Mr. H. A. Webster, Librarian, University of Edin))urgh, for translating a portion of Dlianin's 

 work for me. 



