BUTLER -VND HAFIZ. 169 



ckum is co]ninoii around the hole which a minute boring insect 

 frequently makes in the midrib. Salmon* has found in similar 

 experiments with the mildews {Krijsifhacecti) that '"' green fly " 

 [Afliis] has the same effect as a wound, in weakening the resistance 

 of the plant cells to infection. 



From these experiments it is apparent that there is no essential 

 difference in the ability of the foinis of CoUetotrickum, found on the 

 living midrib of the leaf and on the stem, to attack stems and leaves 

 of sugarcane. Taken in conjunction with their morphological 

 similarity, they nmst be held to be the same fungus. The species 

 appears to be confined to sugarcane. The only other Colletotrichum 

 resembling C. falcatum found widely distributed in India, is C 

 Lineola Corda, which" attacks the leaves of jowar {Androfogon 

 Sorghum) frequently. Morphologically the two species are closely 

 allied, but the jowar fungus does not attack cane leaves. Out of 

 16 inoculations, half on unwounded spots on the midrib, half after 

 scraping oft' the epidermis, none succeeded. Edgerton^ failed to get 

 symptoms of red rot by inoculating sugarcane stems with this 

 species and also with the allied C. cereale. It is probable that 

 many of the new attacks of red rot on the above-ground part of 

 the cane stem, arise as a result of infection by spores blown 

 from the diseased midribs of cane leaves. We do not see how it 

 will ever be possible to avoid this, but it. only gives greater force 

 to the arguments in favour of concentrating attention on the 

 elimination of diseased setts at the time of planting. 



The actual penetration of the fungus into the tissues was studied 

 in the leaf inoculations. Including both leaf and stem forms, 

 altogether 63 inoculations were made on the uninjured midrib, of 

 which 17 succeeded and 46 failed ; while when the epidermis was 

 wounded 20 out of 25 inoculations succeeded. In the successful 

 cases where the epidermis was uninjured, penetration usually oc- 

 curred not directly into the midrib but by superlicial growth of the 

 fungus until the large motor cells lying in groups on either side 



' Salmon, E. S. Cultural experiments -aith " Biologic forms " of tho En/sipJincrrp. 

 I'liil. Trans. Royal Society, Scr. B, Vcl. 107, 1904, p. 112. 

 ■^ Edgerton, C. W., loc. cit. p. 7. 



