232 TRANSACTIONS OF ROYAL SCOTTISH ARBORICULTURAL SOCIETY. 



and although this does not altogether exclude the spruce as a 

 possible host it is somewhat unusual for the same species to 

 attack both hardwoods and conifers. The actual amount of 

 mycelium present in the tissues of the diseased spruce plants 

 is much less than in oak seedlings attacked by R. quercina^ 

 for in the latter Hartig has described the penetration of the 

 cortex by large mycelial strands and the formation of large 

 sclerotium-like masses between the cambium and xylem and in 

 the pith. Although the spruce plants when received had 

 already been seriously injured by the fungus, and would 

 probably have been killed by the disease, yet the number of 

 hyphae in the tissues is comparatively small. This may, 

 however, be due to the conditions under which the plants were 

 kept in the laboratory, conditions which encouraged the 

 development of the fungus, and which would have ultimately 

 brought about the death even of healthy seedlings. The 

 absence of sclerotia in the fungus on the spruce may probably 

 be accounted for in the same way ; sclerotia are usually 

 produced as the result of a check in development brought 

 about by adverse conditions, and no such check would be 

 experienced in the conditions under which the diseased plants 

 were kept. 



There are, however, marked differences from the fungus of 

 the oak disease. The conidiophores in the Scottish specimens 

 do not bear whorls of short branches as in R. quei-cina, but are 

 irregularly branched and of considerable size. The perithecia 

 are produced more abundantly on the spruce, and the ascospores 

 are of decidedly smaller size, being about 20 \x in length, while 

 those of R. quercina are approximately 30 /i. 



These distinctions and other slight differences clearly show 

 that the fungus on the spruce cannot be assigned to R. quercina, 

 and it therefore becomes necessary to consider the other 

 closely allied species of Rosellinia. 



R. necatrix is excluded by the form of its conidiophores, 

 the scarcity of the perithecia and their production only at a 

 very late stage, and the large size of the ascospores, these 

 being about twice the size of those in the Scottish specimens. 



R. aquila has already been mentioned as causing a disease 

 of the mulberry, and this species has been described in 

 Britain on the spruce, but only on the wood of mature 

 trees, which are apparently not affected in any way by 



