literature. Corda included it in his Icones, Vol. 4, fig. 129 (1840) as 

 Sphaeria Htigelii and Saccardo and Massee adopted Corda's specific 

 name probably on the overworked principles of "priority," thinking 

 that 1840 was an earlier date than 1837. 



Cordyceps Robertsii seems frequent in New Zealand. There are 

 five specimens at Paris, and over twenty at Kew and the British 

 Museum, all from New Zealand. I believe it is unknown from Aus- 

 tralia or Tasmania. The variety of Cordyceps Robertsii called var. 

 neglecta can safely be nelgected, as it was based on a specimen that 

 was mashed, which is all the difference as far as I could note. 



CORDYCEPS GUNNII (Figs. 617 and 618). Fertile club, sol- 

 itary, rarely two from same host, 2-3 inches long, 5-10 mm. thick, 

 obtuse, even, with the perithecia imbedded. Stem varying accord- 

 ing to the depth of the larva in the ground, 5-7 mm. thick. Per- 

 ithecia flask-shape, imbedded in the stroma, the mouths only show- 

 ing as minute points on the surface. Ascus with a globose apex. 

 Spores breaking usually into short sections in the ascus. Secondary 

 spores very small, 2*/2 mic. long. 



Cordyceps Gunnii was originally known from Tasmania, but was 

 named from Australian collection of Gunn, who sent Berkeley a long 

 letter on its habits. The host (teste Gray) is the larva of species of 

 the genus Pielus, Lepiodopterous insects, the caterpillars of which live 

 in burrows in the soil. Mr. Gunn wrote that the Cordyceps were 

 found in great abundance in sandy soil. "They are from five to 

 eighteen inches long (according to the depth of the burrow). The 

 stem beneath the surface is white, but the club (2 to 4 inches long) is 

 dark olive black. When fresh the club is from one-third to an inch 

 thick, but one-half inch is the average. Usually one fungus springs 

 from near the head of the larva, sometimes from the back, rarely two 

 from the same larva." 



The type specimen of Cordyceps Gunnii (Fig. 617) is short and 

 thick and has no stipe, but the usual specimen at Kew is more slender 

 (Fig. 618) and has a long stipe. 



Cordyceps Gunnii is known from abundant specimens at Kew 

 and British Museum, all from Australia and Tasmania. There are 

 none from New Zealand. Also there is a single collection at Paris 

 (called Cordyceps Lacroixii) from Japan, which appears to me to be 

 the same species. I have specimens of Cordyceps Gunnii from F. M. 

 Reader, Australia. 



CORDYCEPS HAWKESII (Fig. 619) appears to me, from the account to be 

 but a short-clubbed form of Cordyceps Gunnii. It was named from Tasmania, 

 grew on the same host, and was distinguished by its short club and bearing two 

 fruits near the back or other parts of the larva. Both of these features are recorded 

 by Gunn as occurring exceptionally in his original account of Cordyceps Gunnii. 

 I found no specimen of Cordyceps Hawkesii in either of the museums at London. 



CORDYCEPS TAYLORI (Fig. 621). Fertile club solitary from 

 the host, but branching into several (usually three or four) branches 

 which are again branched near the apex. Perithecia (unknown to 

 me) stated to be superficial. 



6 



