40 



RUST AND MILDEW IN INDIA. 



{Ekamyius). The main diagnostic characters of P. rvbiao are that 

 its uredospore is rounder than that of P. graminis, with several 

 germ-spores irregularly scattered over its surface ; and that its 

 teleutospores are not naked but covered by the host's epidermis, are 

 mostly squared at the free end without appreciable thickening and 

 are exceedingly short-stalked. The teleutospore beds are in Europe 

 surrounded by hair-like structures termed technically paraphyses • 

 but these structures were not present in any of the specimens I 

 have examined in India. We need not consider the characters of 

 the third species (P. coronata), since it is of rare occurrence on 

 cereals, and plays no important part. I have never seen it myself 

 on any Indian cereal, though I have found it frequently on a wild 



grass [Bruchy podium syhoticum). 



The predominant part played by P. graminis in Europe led most 

 observers in other parts of the world to assume that the same 

 species was also the most destructive with them, and this notwith- 

 standing the absence in many places of the barberry, which De 

 J3ary showed to be necessary for the continued existence of the 

 parasite. Thus, throughout the plains of India there is no species 

 of barberry, and I believe there is no barberry in Australia. In 

 the case oi the plains of India, it was necessary to assume that the 

 wheat plants were attacked by the secidiospores of the barberry, 

 which had been wafted to them from enormous distances, as, with 

 the exception of several species of Berberis in the Khasia hills, one 

 on Parasanath (B. asiatica), two on the Nilgiris (B. arutata and B. 

 nepalemis), all the rest are Himalayan. The spores are exceedingly 

 minute and it is quite possible that they may be carried by t°he 

 winds to such immense distances. I found, however, from a careful 

 examination oi the specimens I received through the kindness 

 of friends from the various localities I have already mentioned, 

 that the most prevalent species is not P. gr a minis, but P. rubk,o- 

 vera. Indeed, the only specimen of P. gra minis I received came 

 from Jeypore, where P. rubigo is aho prevalent. Should further 

 observation confirm the opinion which I am now inclined to hold, 

 namely, that the main cause of rust and mildew in India is 

 P. rubigo, we need no longer trouble ourselves with a barberry con- 

 nection. But while we escape this difiSculty we fall into another, 

 namely, that, so far as our present knowledge extends, no ^cidium 

 on any species of the Bonyinea is known. This is in itself, how- 

 ever, no matter of marvel, as few botanists have interested them- 

 selves m this region of their science, and such an iridium has 

 never been looked for. It is more extraordinary that P. rubmo 

 appears to be the most prevalent rust in the Simla region where 

 the barberry ^cidium (in one form at any rate) is most abundant, 

 and where no borage iridium is known to me, although I have 

 searched assiduously for such fungi generally for some years I 

 would not, however, have it understood that such an iridium 



* Since I noticed this, I received the Bulletin issued by Mr. H. L. Bollev 

 already referred to, and I find that in Indiana, also, P. ruligcera appears to be 

 the main cause of rust. * " 



