430 Mr. II. Scott on 



tliey are saturated witli moisture, but there arc no pools, 

 and though swift mountain-streams are numerous at lower 

 levels, they are not present on the peaks; so that on those 

 peaks the only water in which aquatic insects can exist is 

 that which accumulates in hollow leaves, pitchers of Ne- 

 2)enthes, leaf-hases of Pa?ulamis, &c. In speaking of the 

 lily-dwelling Odonate larva? referred to above Dr. Sharp 

 states that the Sandwich Islands are extremely poor in 

 stagnant waters; in large areas of forest the only water that 

 Odonata can find for their larvfe to live in may be small 

 accumulations in plants. 



II. Dominica. — In this island I made two excursions, on 

 both of which I was accompanied and guided by JNIr. Jones, 

 Assistant at the Botanic Station at Roseau, who aided me 

 in collecting, and to whom I am greatly indebted for his 

 kindness. On March 29th we rode to a point above the 

 freshwater lake, whence a view is obtained in two directions — 

 westward down the valleys tow^ards Roseau and the leeward 

 coast, and eastward down on to the windward coast. I am 

 uncertain of the elevation, but it is in the neighbourhood of 

 3000 feet, probably over. Two or three plants of an un- 

 determined bromeliad, growing just low enough to l5e 

 reached, were taken from trees growing at the roadside edge 

 of the luxuriant forest. These contained numbers of 

 Helodiue larvae very similar to, or perhaps identical with, 

 those found in Trinidad : three specimens of a small Staphy- 

 linid beetle, which Mr. Champion tells me is a new species 

 of Stamnodei'iis near to S. optatus Sharp; one worn specimen 

 of a Barid weevil, possibly a species of Nicentrus Casey 

 {teste Champion) ; one (wingless) specimen of a Microvelia, 

 determined by Mr. Distant as identical with the new species 

 {M. insignis) found in Trinidad Bromeliacese ; and some 

 Chironomid larvce. 



On March 30th we visited a piece of virgin forest at an 

 elevation somewhat over 1000 feet in the mountains behind 

 Roseau. Here, in a single epiphytic bromeliad, numbers of 

 the Helodine larv?e were again met with : in my journal I 

 vrote that Hydrophilidse and a Triclioptery^x were also seen, 

 but these are not now forthcoming and were perhaps lost in 

 the hurry of departure ; and a single specimen was captured 

 of a cockroach, determined by Mr. Shelford as a well-known 

 species, Epilumpra conspersa Burm. Some, at any rate, of 

 the species of Epilampra are amphibious, and in the specimen 

 before me the spiracular tubes can be seen projecting from 



1 



