THE WILT DISEASE OF PIGEON-PEA. 9 
instances. Delacroix (5) referred to it the Egyptian cotton wilt, 
and m the same paper suggested that the cause of a disease of 
carnations, previously investigated by him (17), was identical: 
Jaczewski (6) that of sesamum in Turkestan where he has also 
recorded it on cotton and Hibiscus (7) : Malkoff that of cotton in 
Sadovo (8): Zimmermann that of cotton in German East Africa (9) : 
and van Hall that of the same host in the Dutch West Indies (10). 
In these cases, except Zimmerman’s, perithecia were not found 
and the identification was made from conidial stages alone. 
The absence of perithecia has led several observers to exer- 
cise greater caution in the identification of the parasites. Thus 
the “‘ St. Johannis’’ disease of peas in Holland was attributed by 
van Hall (11) to ‘* Fusarium vasinfectum Atk. var. Pisi,’’ a termin- 
ology which while implying specific identity with the conidial 
stages of Smith’s fungus, leaves the genetic connection with 
Neocosmospora open. Similarly Farneti (12) and Montmartini (13) 
used the terms Fusarium niveum Sm. and Fusarium vasinfectum 
Atk. for wilt diseases of cucurbitaceous plants and of Capsicum 
respectively in Italy. 
In the important paper published by Schikorra (14) the name 
Fusarium vasinfectum Atk. var. Pisi van Hall is retained for the 
parasite of the ‘‘St. Johannis’’ diseases of peas, but the author is 
careful to point out (p. 173), that further investigation is necessary 
to decide how far the various Fusaria found within the vascular 
bundles of cultivated plants belong to one or to several different 
species. 
So far as I have been able to ascertain, Neocosmospora vasin- 
fecta Sm. has only been found in its perithecial stage, outside of the 
United States, by Zimmermann in German East Africa until its 
discovery in India in 1907. 
The resemblance between its conidial stages and those of the 
Nectria found on wilted pigeon-pea in Dehra Dun was at once 
evident. Figs. 3, 4 and 5, plate II, of the writer's previous paper 
may be compared with fig. 6, plate V, fig. 9, plate III, and fig. 7, 
