355 



wnll oviposit in the cocoon of Pachymerus gonagra, thongh it is not 

 known whether it vdW develop at the expense of this host, nor whether 

 it will attack it in nature. Probably it will attack other species of 

 Bruchids whenever conditions are favourable in the field. Three 

 distinct larval stages have been observed in H. jyrosopidis and there 

 are doubtless one or two intermediate stages. The full-grown larva 

 spins its cocoon within the pupal chamber of the host and the adult 

 emerges from the seed or pod of the host plant through a circular 

 emergence hole somewhat smaller than that of the host. H. -prosopidis 

 varies in size in accordance with that of its host, those from B. prosopis 

 (the largest of its local hosts) being much larger than those from 

 B. pruinmus. 



A species of Scleroderma bred from Pachymerus gonagra was 

 found to differ from the known Hawaiian species that are endemic 

 and parasitic upon Lepidopterous larvae ; this is believed to be an 

 immigrant species, perhaps from the Orient, and is here described as 

 Scleroderma immigrans, sp. n. This species does not parasitise more 

 than about 10 per cent, of the cocoons of P. gonagra in the locahty 

 examined, and it is not recorded elsewhere in Honolulu upon this 

 host. Five adults of a Eupelmine have been bred from Bruchus 

 prosopis, and the same parasite has been bred from B. pruininus 

 under natural conditions, and also from cocoons of Pachymerus gonagra. 

 This species, originally described as Eiqoelminus swezeyi, Crawl, 

 differs so much from the type of the genus that a new one, Charito- 

 podinus, is created for it. The author takes the opportunity! of 

 describing a new species of this genus from China, Charitopodinus 

 terryi, sp. n. It is doubtful whether C. swezeyi is more than an 

 occasional parasite of Bruchids. 



Pteromalids attacking Bruchids include a species doubtfully referred 

 to Pteromalus calandrae, bred from Bruchus quadrimaculatus infesting 

 pigeon peas in storage ; this species has also been bred experimentally 

 from the Dolichos weevil and from Bruchus chitiensis. The early 

 larval life is spent as an internal parasite of the Bruchid larva ; when 

 nearing maturity the Pteromahd larva emerges from its host and 

 completes its development externally. Another undetermined Ptero- 

 malid has been bred from Bruchus pruininus in seeds of Sesbania 

 sesban hanging in partly opened, pods on the tree. 



A mite, Pediculoides ventricosus, is found to cause much destruction 

 among all stages of many Bruchids, as well as of their parasites. These 

 mites affect the beetles more generally in some seeds than in others, 

 according to whether the texture of the seed permits ready entrance 

 or not. Any introduction of the larval parasites of Bruchids would 

 have to be carried on with particular care in order to reduce the 

 attacks of this mite upon them. 



This paper concludes with observations on the o^dposition habits 

 of Bruchids with special reference to the fact that this is often effected 

 in totally unsuitable places, while species such as Bruchus pruininus 

 deposit eggs on Ipomoea seeds, castor beans, and seeds of Cassia 

 glauca, in which the larvae cannot survive, or on indigo seeds which 

 produce only an undersized adult. This characteristic may enable 

 the species to live on unusual food when its own food-plant fails, and 

 may also permit a wdder dispersal of the insect. 



Certain Bruchids are remarkable for attacking the seeds of palms ; 

 (C485) -2 



