214 



Experiment Station igrounds, 33.3% were found to contain in- 

 sect remains; leaf-hopper fragments were found in 15-5^5 

 15.5% contained vegetation; 40% empty. 



Examination of their crops does not prove to what extent 

 they eat leaf-hoppers, for they do not always swallow the hard 

 parts of these. Often in the breeding cage they were seen 

 to chew up leaf-hoppers, and reject them after sucking the 

 blood and juices. However, from these examinations, and field 

 and insectary observations, it is certain that this grasshopper is 

 very extensively insectivorous, and that when present in cane- 

 lields, leaf-hoppers form the chief part of its diet, and it there- 

 fore s'hould be classed as a beneficial insect, and a highly valu- 

 able enemy of the leaf-hopper. 



In March, 1905, parasitized eggs of this grasshopper were dis- 

 covered in cane on H. S. P. A. Experiment Station grounds, 

 and a pale yellow Chalcid parasite was bred from them. The 

 number bred, ranged from 9 to 14 per Xiphidium egg, and in 

 one instance a whole cluster of 10 eggs had been parasitized; in 

 another case 13 eggs were parasitized. All the parasites in an 

 egg usually issued from the same hole, (PI. XIV, fig. 11), the 

 first one to mature probably gnawing a^ hole out, and the 

 others taking advantage of the opening already made. This 

 tiny round hole is often made through the enclosing sheath of 

 the cane leaf as well, where the eggs are quite tightly enclosed 

 by the latter. Parasitized eggs were examined from' which the 

 parasites had not yet emerged. They are a bluish black in color. 

 Larvae were found in some, and pupae in others. The pupae 

 were not enclosed in individual cocoons. 



Adult parasites were placed with fresh grasshopper eggs, in 

 which they were soon seen ovipositing. In one case adult para- 

 sites emerged in 25 days; in others 22-24 days; in one 20 days; 

 another 31 days; giving a range of from 20 to 31 days for the 

 life cycle. 



The shortness of life cycle compared with the length of life 

 cycle of the host (3 months) gives the parasite the advantage, 

 and it is likely that this grasshopper may become reduced in 

 numbers more rapidly than it has increased since its appearance 

 a few years ago. In fact, it has been reported as rapidly dis- 

 appearing recently in the Kohala District, Ewa Plantation, and 

 some others. This is undoubtedly due to this parasite, though 

 its presence there has not been ascertained. The grasshoppers 



