FAM. MEMBRACID^ 17 



each thoracic segment is broad, smooth, and, with the exception of the pronotum, simple. The 

 pleuron is narrow, irregular, and more or less complicated, the sclerites are incHned to be twisted from 

 a nonnal position. The sternum is broad, much sculptured, and indistinctly sutured. 



The Prothopax. — No evidence of cervical sclerites has been found. The only suggestion 

 of such structures is a slight thickening of the connecting membrane in the gular region, which in certain 

 species is of sufficient size to warrant attention. On the whole the inembranous connection between 

 liead and prothorax is remarkably thin and easily ruptured, and shows nothing that could be con- 

 sidered as intersegmentalia or could represent the « microthorax » of Verhoeff (1902). 



The notum of the prothorax shows so mucli variation throughout the family that no general 

 discussion of it can be attempted. The peculiarities of this region represent by far the most striking 

 and easily recognized characters of the Membracidae. This part of the prothorax is usually expanded 

 into a more or less irregular plate, which covers the entire meso- and metanotum, often the entire 

 thorax, and in some cases the abdomen as well, and bears on its surface a wide variety of processes 

 extending to form most grotesque and bizarre structures. A discussion of such variations would be 

 merely an endless catalogue, and is of course not to be attempted. Apparently the pronotal structures 

 have no anatomical significance and are merely hollow extensions of the chitinized wall, raised high 

 above the basal membrane which represents the normal body outline. Moreover, extended experiments 

 clearly indicate that these structures are not involved in any way with any of the physiological processes 

 of the insect. In fact, in many instances, the insects seem to fare better without them than with them. 

 But whether or not they have any functional value, these pronotal developments are a boon to the 

 writer of descriptions since they lend themselves so well to diagnosis and are apparently quite constant, 

 both as generic and specific characters, and some apply, at least as secondary characters, to each sub- 

 family. It may be noted in this respect that the j)ronotum tends to develop in four principal directions 



— posteriorly, anteriorly, dorsally and from the humeral angles. These four great types of development 

 may be found in various stages of enlargement throughout the family, and on them are based many 

 attempts of subdivision into subfamilies, tribes and genera. Modifications and combinations of these 

 types are of course common, and in sonie groups it is difficult to decide which type is dominant. 



By far the commonest of these types is the development posteriorly, to cover the meso- and 

 metanotum and often the entire body of the insect. This posterior extension is found in so large a 

 proportion of the forms that it appears to be a sort of foundation structure on which the other types 

 of development are built, and is apparently one of the most generalized of the prothoracic processes. 

 It may vary from a perfectly simple short prong to a long ornate projection often branched, extrava- 

 gantly decorated with barbs, spines, bulbs, and ridges. So constant and so important is this posterior 

 process that it has been made the character on which the subf. Cenlrotitia is separated. All forms that 

 have the posterior process wanting or so poorly developed that the scutellum is distinct — and it 

 would seem that the development of the scutellum increases as that of the posterior process decreases 



— have been placed in this subfamily, which as a result has received a rather heterogeneous collection 

 of genera. In generic and specific diagnoses the pronotal structures have been more generally used 

 than any other characters shown in the family. This is true for the posterior process, the size and 

 shape of the humeral angles, the character of the suprahumeral horns, the structure of the dorsal 

 humps and spines and the extent of the median carina. This is to be expected, from the fact that 

 these structures are very prominent and quickly noted. Moreover, they are on the whole reliable and 

 of much value. 



It would be impiacticable to attempt to indicate the great number of ways in which these struc- 

 tures may vary. It would seem, however, from the examination of all of the genera in the family and 



