Hill — On Cordiceps robertsii. 399 



common in the uplands where the soil is porous and 

 pumiceous, as between Napier and the volcanic country, and 

 the specimens vary remarkably in size and general appear- 

 ance. They are not obtainable at all times of the year. I 

 have gathered them at Tarawera and Te Haroto, in the bush, 

 during October, November, and December, and they were 

 being sold fresh and in good condition by the native children 

 along the railway-line leading to Eotorua late in March two 

 years ago. 



The specimens vary from If in. to 3^- in. in length, and the 

 shoots or stems vary from 3 in. to as much as 15 in. I 

 happen to have available for reference two collections of the 

 vegetable caterpillar, one of my own and one belonging to my 

 friend Mr. A. Hamilton. They were collected from different 

 districts, and a careful inspection of them seems to point to 

 the fact that a special study of the vegetable caterpillar will 

 provide some valuable information in the direction suggested 

 by the late W. M. Maskell. 



It is not yet known what grub it is that inhumes itself in 

 the earth, but I do not think it is the Hepialus, as the seg- 

 mentation is different, and the mandibles do not agree. I 

 am endeavouring to obtain specimens of the live caterpillar 

 from the natives, who say that it can be got at certain times 

 of the year ; but, unfortunately, it is difficult to get a native to 

 proceed into the bush for the mere purpose of gathering a few 

 caterpillars. 



In order to show that the prevailing idea about the vege- 

 table caterpillar is an incorrect one, I have photographed a 

 number of specimens, each being different from the others : 

 (1) Caterpillar with a single stem ; (2) caterpillar with two 

 stems at the same end ; (3) caterpillar with bifurcated growth 

 from the same end ; (4) caterpillar with much- branched stem, 

 like a stag's horn ; (5) caterpillar with two stems united 2 in. 

 or so from the head, followed by a bifurcation, one with and 

 one without spines ; (6) caterpillar with two stems, one at 

 the head and one at the tail. These examples suffice to show 

 that the caterpillar, when it inhumes itself in the ground, is 

 affected by the spores at a different period in its movements, 

 and there seems to me not a shadow of doubt that the cater- 

 pillar moves head foremost into the earth. I have quite a 

 number of specimens where it is evident the fungus began to 

 shoot as the caterpillar was head downwards, others where 

 the caterpillar was seemingly preparing to leave the earth 

 when the fungus began its deadly work, and absorbed the 

 whole of the life's-blood of the caterpillar in the nourishment 

 and maintenance of its own organism. The specimen with a 

 shoot at either end is a proof in itself that the animal was in 

 an almost horizontal position when it was attacked by two of 



