88 Messrs. C. B. Williams and P. A. Buxton on 



forming a protective shield, or plate, which ends dorsally 

 and ventrally in a pointed prolongation (PI. VII, fig. la). 

 The sides of this are reticulated for a short distance, and 

 it is this reticulation that was apparently mistaken by 

 Pagenstecher (1864) for the covering of the eyes (" augen- 

 flecken "). This plate may possibly be used in breaking 

 open the egg. 



On the ventral surface at the posterior end are two 

 pairs of small papillae, one pair on the hind margin of 

 the ninth sternite and the other beneath the sides of the 

 posterior margin of the tenth tergite. The posterior pair (c) 

 secrete the two threads by which the larva hangs from 

 the ootheca. These threads frequently unite a short 

 distance behind the larva, forming a flat band. Brongniart 

 (1882, p. 451) states that one thread is attached to each 

 end of the eggshell. This, however, is not so, as in sections 

 of ootheca which expose empty eggshells, both threads 

 can be seen entering the shell, and are attached to its 

 interior surface near to the posterior end. 



It is remarkable that we were able to draw out a con- 

 siderable length of thread (2-3 mm.) from the papillae 

 of a cast skin. Whether the whole thread-forming ap- 

 paratus is cast with the skin we have not been able to 

 decide, but it should be noted in this connection that at 

 no other stage in the life of the Mantis is any thread or 

 silk produced. According to Giardina, Pawlowa (1896) 

 studied the development of the thread and found that it 

 was formed from a mass of cells arranged in a single 

 spiral series which take the form of a cellular filament. 

 This undergoes morphological and chemical change until 

 it loses all cellular structure and is reduced to a fine but 

 resisting thread. We have been unable to consult 

 Pawlowa's or ginal work. 



Pagenstecher (1864, p. 13) and de Saussure (1872, p. 223) 

 describe the presence of a number of backwardly directed 

 spines on the abdomen of the just-hatched larva, by means 

 of which they are enabled to work their way up the passage 

 from the egg to the exterior, according to the latter " de 

 la meme maniere qu'un epi de seigle, a l'aide de barbes a 

 ergots, peut cheminer sur un morceau de drap soumis a 

 des vibrations." Brongniart (1882, p. 450) denies this, 

 and states that the spines on the body are confined to the 

 cerci, and also remarks that the spines on the legs assist 

 in the exit. Recent authors have unfortunately tended 



