juneis. I9IS Sugav-Cane Root-BoHng WecvUs 257 



THE IMMATURE STAGES 



A sketch of an egg is given in Plate XXXVIII, figure i, F. 



The larvae of weevils are most readily distinguished one from another 

 by the folds of the body, the shape and arrangement of the spiracles, 

 the head, and the caudal segment. Sketches of these parts are there- 

 fore given to assist in the identification of larvae of Diaprepes spengleri. 

 These drawings have been made with great accuracy with the intention 

 of bringing out the systematic characters (PI. XXXVII, fig. 2, and PI. 

 XXXVIII, fig. I, A-E). 



In the pupae of weevils we find even better characters for separating 

 species. The general form of the pupa and the arrangement of the 

 ventral parts is of considerable importance (PI. XXXVIII, fig. 2). The 

 mouth parts of this species are very interesting and have been illustrated 

 in Plate XXXVIII, figure 2, B. The arrangement of the tubercles on 

 the dorsal segments is important, especially in the scutellar region (PI. 

 XXXVIII, fig. 2, C),and on the apical segments (PI. XXXVIII, fig. 2, E). 

 But more important than any other characters are the structures of 

 the last three or four ventral segments (PI. XXXVIII, fig. 2, D). It is 

 hardly necessary to add much of a descriptive nature to the figures given. 

 The spiracles of the abdomen are dark and very plain on the first to 

 fifth segments, and very inconspicuous on the sixth to eighth segments. 

 The thoracic spiracle is elongate and located between the prothorax and 

 mesothorax. 



VARIETIES OF DIAPREPES SPENGLERI 



Merely for the convenience of designation and to retain old, well- 

 known names the species Diaprepes spengleri has been arbitrarily arranged 

 into varieties by the writer. 



The variety which is presumably nearest the parent variety, Diaprepes 

 spengleri marginatus Olivier, is only at hand from St. Croix. The next 

 step in the progression of the species has been called D. spengleri comma 

 Boheman, and ranges from the Dominican Republic and Porto Rico to 

 Dominica; in other words occurring in both directions from the home 

 of D. spengleri marginatus , with nearest approach to this form in a speci- 

 men from Dominica. The next form is merely an intermediate and is 

 found in the collection only from Porto Rico. This form most nearly 

 answers the description of typical D. spengleri Linnaeus. The fourth 

 variety, D. spengleri ahhreviatus Olivier, is at hand from Porto Rico, 

 Montserrat, Dominica, and Barbados. There seem to be two trends of 

 modification from this. The fifth variety, D. spengleri denvdatus, n. var., 

 is an extreme from a trend found in the Porto Rican material, and is at 

 hand only from Guadeloupe. The sixth variety, D. spengleri festivus 

 Fabricius, is from the branch of the fourth found in Dominica and Bar- 

 bados, and is at hand from Barbados and St. Vincent. 



Owing to the remarkable differences displayed in the species, a more 

 or less detailed study follows to show why the writer has considered 



