128 The South Australian Naturalist. 



aquatic animals are captui'ed by them. On account of their 

 predatory liabit, it Avas found necessary to isolate each indi- 

 vidual in a separate jar, for otherwise the smallest example:< 

 were desti-oyed hy tlieir larii^er brethren. 



As is well-known, bugs haA^e an ** incomplete" metamor- 

 phosis; the nymph of our back-swimmer moults its skin five 

 times before it finally attains maturity, an increase in size 

 and some developmental chanj^es being apparent after eacli 

 ecdysis. The ti-ansition, unlike that of insects Avhich undergo 

 a ''complete" metaiiioi-phosis, is gradual. It may be said 

 that when, for instance, the bug attains the second stage or 

 instar, the thiixl instar nymph begins to form inside the skin 

 of the second instar nymph ; then, when the stage is comple- 

 ted, the skin is split along the back and the nymph emerges, 

 immediately expands to the proportions of the third instar. 

 and so on until the fifth and final moult. The ventral carina 

 and respiratory gutters are present in the second instar, and 

 the first indications of Aving-pads appear in the next, or third 

 stage. The last nymphal stage is reached about a month after 

 hatching, and as these fifth instai- nymphs near the final moult, 

 the characters of the developing adult may be seen throng a 

 the semi-transparent integument by means of strong trans- 

 mitted light under the microscope. The stridulatory apparatus 

 of the male is thus A'isible, and other structures of both sexes 

 may be similarly discerned, so that it is possible to determin*^ 

 Avith ease the sex of advanced indiA'iduals of this instar. When 

 the metamorphosis is completed, the integument of the adult 

 is at first soft, and little color is apparent on the underside 

 of the abdomen. 



In aquaria each nymph captured on an aA^erage 200 mos- 

 quito larvae betAveen the second and last moults, and a laying 

 female consumed 70 large larvae and pupae in one month. 

 Food is held AA'ith the bristle-edged anterior and intermediate 

 legs, the victim being dexterously turned about as the beak is 

 applied to fresh portions. 



As an experiment a laying female Avas isolated in a vessel 

 containing plants of Italian Ribbon-Aveed (Vallisneria spiralis), 

 but no other A-egetation. The thin, tape-like leaA^es of this 

 Avater-plant are not suited for the reception of eggs, and in a 

 Aveek or so many unoccupied perforations indicated abortive 

 attempts to prepare cavities in AAdiich to conceal the oA'a. Eggs 

 Avere, hoAvever, pushed into certain of the slits cut in the 

 leaA'es, but Avere, of course, largely exposed, 



