MORPHOLOGY OF THE ALEYRODID^. 15 



appearance, for here the ring of spinnerets is not extended into rod- 

 shaped structures, but is seen only at the base. The outer cup is very 

 shallow and the central chitinous process is split into a number of 

 sections. It would appear that in Parcdeyrodes the spinnerets and 

 outer cup have become shortened at the same time with the media 

 and radiuSj. The compound pores of the forms belonging to the 

 subgenus Metaleurodlous are of an intermediate nature. The spin- 

 nerets are reduced in minima and cardini in a manner comparable to 

 those of Paraleyrodes^ but the central process is entire. In aUissimus 

 besides the reduction of the spinnerets, the central process is split into 

 three divisions. In these forms, however, the lingula is short and 

 generally included, unlike the typical forms of Alenrodicus or 

 Paraleyrodes. Unfortunately the adult of cdtissimus has never been 

 secured as, from the pupa, it appears to be a very interesting form; 

 but Avhether it is nearer Parcdeyrodes or Aleurodicus cnn be decided 

 only by an examination of all the stages. 



TJESPIRATORY SYSTEM. 



The respiratory system in Aleyrodida seems to be more fully devel- 

 oped in the larval stages than in the adults. In some forms, as in 

 those belonging to the citri group, there are in the immature stages 

 special breathing folds developed which are often indicated on the 

 dorsal surface by sculpturing and the outer extremity of each is 

 guarded by stellate or other serrations on the margin of the case. 

 The folds are three in number, there being a latero-thoracic pair and 

 a caudal fold, this latter extending from the vasiform orifice caudad 

 to the margin and forming a Y-shaped structure which extends to the 

 posterior thoracic pair of spiracles. In some species, even when the 

 thoracic folds are not noticeable, the caudal fold will be long and 

 deep. The courses of the various trachea3 have been worked out b}'^ 

 AVood worth for Aleyrodes citri^'^ and the observations of more recent 

 writers on the subject, as well as those made by ourselves, only sub- 

 stantiate his findings. 



In the adult the system follows the same plan as in the larva. 

 There are the same number (four pairs) of spiracles present, one pair 

 of which is abdominal and situated near the vasiform orifice. The 

 trunks and girdle are, however, much more reduced in the mature 

 insect. 



METAMORPHOSIS. 



The metamorphosis in this family has always been an interesting 

 and problematical subject of study. During the later larval stages 

 the antenna^, mouthparts, and legs become, to a certain extent, atro- 

 phied. In the pupal stage, however, they appear to be, in some cases 

 at least, functional. In Dialeurodiou^ cocL-erellii, for example, there 



