44 ARKIV FÖR ZOOLOGI. BAND 8. NIO 9. 



Finally, in Phyllocnistis, where the labium and hypo- 

 pharynx have attained the climax of development, it is pos- 

 sible with oil-immersion to discern a pair of small lines, which 

 may be the two hairs found in Lithocolletis. At the point 

 where the salivary duct ends, we notice two pairs of small, 

 pointed appendages, which are probably the remnants of the 

 spinneret and the labial palpi. 



As pointed out above, it is very probable that the* end 

 of the salivary duct marks the anterior margin of the labium, 

 which is visible as a curved line, not very pronounced in 

 Ornix, but very distinct in Gracilaria; and it is not improb- 

 able that the labium with a thin edge overlaps the posterior 

 margin of the hypopharynx. This is, however, very difficult 

 to ascertain. At any råte, in Gracilaria we notice two fine 

 lines running parallel to the former, without either meeting 

 the median line or the lateral margin; in Lithocolletis this line 

 is more developed, and in Phyllocnistis it reaches the margin. 



The hypopharynx, which is characterized by being den- 

 sely clothed with small hairs, shows a gradual increase in 

 size in Ornix— Gracilaria — Lithocolletis— Phyllocnistis. 



9. On the organisation of the låter instars of 



the sap-feeders. 



The length of time during which the larvae of Gracilaria, 

 Ornix, Lithocolletis and Phyllocnistis are sap-feeders, differs 

 to all appearance in the different genera, but this question 

 has not been thoroughly enough investigated, to enable us to 

 give any reliable data, except in Phyllocnistis suffusella where 

 through the researches of Sich and others, which I have been 

 able to verify, we know that all instars, except the last one, 

 are of the sap-feeding type and which genus represen ts the 

 extreme of this remarkable modification. 



In the other genera at least the two last instars are not 

 sap-feeders, but have a different mode of feeding. 



Those of Gracilaria roll a leaf together after the mari- 

 ner of a Tortricid-larvsi, while in Ornix guttea, for instance, 

 they turn down the edge of the leaf, and fästen it to the 

 lower surface, thus forming a narrow tube in which they 

 devour the parenchyma. 



