16 



ARKIV FÖR ZOOOGI. BAND 7. NIO 28. 



hairs instead of the usual 3 pairs. At the same time the 

 shield which occupies the place between the sternal and epi- 

 gynial shelds is much smaller than in those genera where the 

 sternal shield has only 3 pairs, and, furtbermore, it bas no 

 hairs. 



In otber genera, e. g. Holostasjns (Fig. 24), where the 

 sternal shield bas 3 pairs of hairs, there are, in the space 

 between that shield and the epigynial one, two pairs of 

 shields, a small rounded one, with a hair on it, and another 

 narrow, curved, bairless one which is undoubtedly homolo- 

 guous to the bairless one of Gamasellus and others, and is 

 an endopodal shield. 



If we combine these facts, we shall be able to draw two 

 conclusions. The first, that an essential feature of the me- 

 tasternal shield is the presence of a hair on it; the second, 

 that the endopodal shields are not provided with any hairs. 

 If, in consequence, we find 4 pairs of hairs on the sternal 

 shield, tbis is a conclusive proof that the shield is morpho- 



logieally composed of the coa- 

 lesced sternal and metasternal 

 shields. 



A f urther proof of the hairs 

 on these shields not being endo- 

 podal, in spite of their lateral 

 position, is furnished by the ge- 

 nus HalolcelapsB^Rj^. & Trouess 

 (Fig. 25), where the endopodal 

 shields are not fused with the 

 sternal shield, and where the 

 hairs are inserted in the latter. 

 Based on the different po- 

 sition of the metasternal hairs, 

 we are thus able to ascertain 

 that the metasternal shields may 

 either be free (as for example 

 in H olöstas pis and Trachytes), 

 fused with the corresponding 

 endopodal shields (e. g. Perga- 

 masus and others) or with the 

 sternal shield (e. g. Protolcelaps and others). There are, how- 

 ever, genera in which the female epigynial shield has no hairs, 



Fig. 25. 



Halolcelaps. 9, ventral vievv 

 (after Berlese). 



