60 THE REPOET OF THE No. 36 



on the Tall Meadow Eue {Tiialictniui polyganium) , where it feeds on the white 

 flower clusters. These larvae will drop to the ground when one is four or five 

 feet from the plant on which they are feeding. When they reach the ground 

 they curl themselves up in a ball. The adult has not been <bred. In striking 

 contrast to this larva is that of Dimorphopteryx, which feeds resting upon the 

 l^pper surface of the leaves of birch, linden, maple, black oak, and shadbush. 

 It is impossible to shake these larvae loose and difficult sometimes to pull them 

 loose with the forceps. 



The larva^ of certain genera and subfamilies of Tenthredinidae are entirely 

 different in appearance during their last larval period; white larvae may become 

 spotted, the spotted white or green, and the spiny spineless. When the larvae are 

 through feeding, they cast their skin, which may take place either before or after 

 leaving the host plant. It is the period during which they are preparing for 

 pupation. For one who has not observed the differences found at this time it is 

 difficult to appreciate the change that takes place. My first experience was 

 with a larva that was glazed white when I examined it in the afternoon; the 

 next morning the breeding cage contained almost entirely larvae that were white 

 with small irregular black patches. This stage has been named by Dr. H. 

 G. Dyar, who has done more than all the other American workers together in 

 elucidating the transformations of the Tenthredinoidea, the ultimate stage. The 

 assumption of an ultimate stage does not seem to be peculiar to subfamilies or 

 genera. The modifications in color and structure will be discussed more fully 

 later. 



There is considerable variation in the method adopted by the larva? in pre- 

 paring for pupation. The Xyelidae and Pamphiliidae transform in cells in the 

 ground. The species of Emphytinffi and Selandriinae for the most part bore into 

 rotten wood ; this is particularly true of the species that feed on different species 

 of ferns, of which there is a considerable num'ber. These larvse will transform 

 in absolutely no other way. They will travel round and round a breeding cage 

 for days searching for the rotten wood, and if placed upon a piece, will be out 

 of sight within it in a few minutes. They cut a smooth, straight tunnel of the 

 same diameter as their bodies and plug the open end with the dust from their 

 cuttings. The tunnel is usually not lined with silk, althoug'h a few species prepare 

 a very slight silken lining. All the species of Macr-emphytus, Empliytus, Strongy- 

 logastroidea, Strongylogaster, and Ta.roniis transform in this way, also a few 

 nematids. The larvse of the Blennocampina?, transform m cells in the ground. 

 The Lopherinas, Dolerinffi, Phyllotominae, Tenthredininae, Hoplocampinte, Hylo- 

 tominse, and most Nematinae transform in cocoons Just beneath the surface of 

 the ground and in the breeding cage will form their cocoons, if no soil or trash 

 is provided, on the bottom or side of the jar. Some Lophyrins, also Cimhex 

 and Trichiosoma, sometimes attach their cocoons to the host plant. The cocoons 

 of all these groups except the Hylotominw are brownish or blackish in color and 

 dense in structure, the cocoons of the Hylotomina? on the other hand are white 

 and lace-like and appear much larger in size in proportion to the size of the 

 adult. The cocoons of the Acordulecerinae are minute, dense, and opaque white 

 in color, and are probably placed just beneath the surface of the ground. They 

 were made among the debris of the food material in the bottom of the cage. 

 The Scolioneurinae and Fenusinse leave their mines and enter the ground, where 

 a part of the species at least form cocoons. The larv» of the Xiphydriidae, 

 Siricidae, Cephidae, and Oryssidae transform in their tunnels, some, if not all, 



