192 THE ZOOLOGIST. 



structs a beautiful little case of pieces of leaves. Miiller found 

 an allied larva with a similar " house " inhabiting rivulets in 

 the same region. The pupa of the rivulet-dwelling species 

 (which he referred also to the genus Phylloicus) agrees with 

 many other trichopterous pupae in having the middle legs fringed 

 with hairs to form a swimming-organ, which aids it in reaching 

 the surface on its emergence from its cocoon. The pupa of 

 P. bromeliarum, however, is devoid of these fringes, which may 

 be correlated with the fact that it probably has no need to swim 

 in order to reach the surface in the narrow spaces of the plant.* 

 One of the' most interesting bromelicolous animals is an 

 Agrionid Dragonfly, Mecistogaster modestus, Selys. The insects 

 of this genus are characterized by the extremely exaggerated 

 length and slenderness of the abdomen, which gives them a 

 highly remarkable appearance. Till recently nothing was known 

 of their early stages or breeding-places, but in 1900 it was 

 suggested that they might breed in bromelias. In 1908 specimens 

 were actually bred by Knab from larva? found in these plants in 

 Mexico, and Calvert in 1909-10 worked out the life-history of the 

 insect in Costa Rica. The abdomen of the larva is not dispro- 

 portionately long, and the extreme length in the imago is 

 acquired by a very rapid extension during the hours immediately 

 following emergence from the nymph-skin. Thus it may be 

 that the great length of the abdomen in the adult is a secondarily 

 acquired character, and Calvert has suggested that it is possibly 

 an adaptation for ovipositing between the leaves of bromelias in 

 spaces which are too narrow and deep to admit of the female 

 climbing down.t 



* I am indebted to Mr. K. J. Morton for referring me to the literature 

 on this matter. Miiller described and figured these larvae in a paper " Sobre 

 as casas construidas pelas larvas de insectos trichopteros da Provincia de 

 Santa Catharina," Arch. Mus. Nac. Rio Janeiro, iii., 1878. He mentions 

 P. bromeliarum on pp. 114, 115, without naming it, and again (in the 

 Supplement) on pp. 131, 132, this time giving it its name; the case of the 

 larva is figured on pi. 9, fig. 17. In the following year (1879; he published a 

 paper in English in the Trans. Ent. Soc. London, and on p. 137 of this 

 volume he refers to the same larva, but without giving it a name. It is also 

 referred to by McLachlan on p. vii. of Proc. Ent. Soc. for that year. 



| Calvert, ' Entomological News ' (Philadelphia), vol. xxii., 1911, pp. 402- 

 411 and 449-460, pis. xviii.-sis. 



