92 THE CANADIAN ENTOMO'LOGIST 



The subfamily Disapidinae has been the special recipient of MacGillivray's 

 attention and occupies 249 of the 465 pages of text in the book. Throughout 

 this subfamily the -torm of genus making has raged, reaching its climax in the 

 Aspidiotini, and of the old genera there remain nothing but fragments. I count 

 116 new genera in the sub-family. To attempt to analyze these genera in 

 anything short of another book is impossible and I shall content myself here 

 with discussing only a few of the most remarkable results and some of the 

 most obvious errors. Yet the commission of such errors as some of these are 

 is sufficient to cast doubt upon the value of MacGillivray's conclusions in gen- 

 eral. It is conclusive evidence that he has not attained to that knowledge and 

 understanding of the group that should have preceded any such wholesale re- 

 arrangement as he has undertaken. 



In the general discussion of this subfamily I note two errors that may 

 be of some iivportance. On page 218 it is stated that in the first stage 

 nymphs the distal segment of the antennae "is long and constricted and appears 

 as if composed of several segments." This is true of only a part of the group, 

 for in some species the distal segment is short and not at all annulated. On page 

 220 it is said that when the insects of the second stage molt "there is no 

 variation in the way in which the cuticle ruptures." However, in some forms 

 such as certain species of the genus Odonaspis (as previously understood), the 

 ventral skin separates entirely from the dorsal and is incorporated in the vent- 

 ral scale. / 



I am not entirely in accord with MacGillivray's arrangement of the tribes 

 in this subfamily, although this is not original with him. I consider that the 

 various groups of species in which the adult remains enclosed within the 

 second exuvia are, with certain exceptions, derivatives of forms that are in- 

 cluded in the various other tribes and that separate tribes for such groups as the 

 Leucaspidini and Fioriniini tend merely to ob'scure their real relationships. 

 Furthermore I consider the distinction between the tribes Lepidosaphini and 

 Disapidini to be entirely artificial. The genus Anccpaspi<, v.diicli MacGillivray 

 attaches to the Aspidiotini I consider to represent an independent group. 



In many cases tliroughout the subfamily MacGillivray has separated 

 genera on the basis of the presence or absence of the paragenital pores, the 

 "genacerores." That the consistent following of this practice results in artifi- 

 cial groups seems to me evident as for instance in tlie cn=e of the genera 

 Lincaspis and Cupidaspis, the types of which are scarcely separable specifically 

 except for this difference. Such splitting as this may be convenient but it does 

 not express the relationships of the forms involved. 



The peculiar combinations ol^tained by MacGillivray's methods begin 

 to appear in the tribe Parlatoriini where Parlatoria chincnsis Marlatt and P. pyri 

 Marlatt are referred to the genus Crypto parlatoria with C. Icucaspis Lindinger 

 as type. If specimens before me as C. Icucaspis be correctly determined such an 

 arrangement is quite untenable. 



