1. Ustilago Panici miliacei occurs very abundantly on Panicum miliaceum in Hokkaido (Yezo). 2. Usually the whole inflorescence is changed into the smutpustule bearing sometimes an undeveloped bract on or near its apex (Fig. I). 3. The internode immediately below the inflorescence is often affected and forms the lower portion of the smutpustule; in this case a more or less well developed leaf is borne on the smutpustule (Figs. II and III). Partially smutted inflorescences are also very rarely found (Fig. IV and V). 4. The coating of the smutpustule consists of the epidermis of the host and a layer of sterile hyphæ, which is in close contact with the inner surface of the epidermis (Fig. VI). 5. The smutpustule is traversed lodgitudinally by the vascular bundles of the host and the strands of sterile hyphæ, spoken of by Dietel (1) as Zellcomplexe in the case of Ustilago Ischæmi Fckl. 6. The hyphæ of these strands, which are similar to, but distinctly bigger than, those of the coating, are close-septate and become divided up into round or oblong simple cells, when the spores are ripe (Fig. VIII). 7. The spores are produced at the end of strongly gelatinized hyphal branches which compactly intertwine forming a large ball. (Fig. IX and XI). 8. The spore-balls are surrounded while they are young by an investment consisting of gelatinous sterile hyphæ (Figs. VI, s b and X). At maturity, these balls which then measure 45-105μ×45-85μ in size lose their investment and become easily separable. 9. The spore-balls are formed in more or less concentric rings around each vascular bundle, and the ripening proceeds centripetally toward the bundle (Fig. XII). 10. As can be gathered up from the above account, the present species is evidently referable to the genus Sorosporium, and should bear the name Sorosporium Panici miliacei (Pers.). 11. Resemblance of the fungus in question to Sorosporium Syntherismæ (Schwn.) Farlow was pointed out by Pammel(1) in 1892, when the latter species was still being retained in the genus Ustilago. 12. The hyphal stratum inside the epidermis of the smutpustule is well developed in Sorosporium Syntherismæ and also in S. Everhortii Ell. et Gall., as far as my observation is concerned. This is also the case with S. Ehrenbergii Kühn according to Dietel. (2) In the above-named two species of Sorosporium, however, the hyphal strands are wanting in the smutpustule, at least, in the specimens which I have examined.