119 



Mere casual observation of many strains of Helminthosporium showed 

 that the point of maximum diameter was usually near the basal region of 

 the conidium, occasionally near the middle region, while in extremely rare 

 cases it was near the apical quarter, the ratio of these cases being for H. 

 No. 1 about 30:14:4. A more accurate determination of what may be 

 termed the longitudinal eccentricity of the conidium that is the range 

 of variation in the position of the line of greatest diameter (a a ' , Diagr. 

 I IV) may be made by measuring (along the longitudinal axis) the dis- 

 tance from the base of the conidium to the intersection of the line a a! 

 with the longitudinal axis. This distance divided by the total length of 

 the conidium may be known as its coefficient of longitudinal eccentricity. 

 This coefficient for H. No. 1, based on 65 conidia taken at random, was 

 found by the above method to be .43 0. In other terms the point of 

 maximum diameter was distant from the base of the conidium 43% of the 

 total length of the conidium. Bakke (6) says that the conidia of H. te- 

 res are widest at the middle. The coefficient of longitudinal eccentricity 

 based on 11 conidia of H. No. 1 which were of typical subcylindrical ap- 

 pearance (approaching that shown in Diagr. IV) was .45 as contrasted with 

 a coefficient of .43 for 11 conidia of elliptical appearance (Diagr. I). Co- 

 efficients of longitudinal eccentricity for H. Nos. 5, 20, and 4 of subcylin- 

 drical shape, were respectively .35, .39, and .37, showing that in these 

 forms the point of maximum diameter is slightly nearer the base than it is 

 in H. No. 1. None of the conidia of H. No. 1 was truly cylindrical, 

 that is, the sides were not parallel for any appreciable distance. Many 

 were subcylindrical, the form approaching that shown in Diagram IV. Of 

 65 conidia taken at random 81%+ of the conidia were elliptical; 17% + 

 subcylindrical; and 1% otherwise. 



To secure a coefficient which would indicate with some degree of ac- 

 curacy the curvature of the conidial wall (as from point a to point y, Diagr. 



xy 

 I IV) determinations were made of the ratio (Diagr. I IV). The 



> 



line cd was tangential to the surface of the conidium at the point of maxi- 

 mum diameter, and was parallel to the longitudinal axis of the conidium, 

 the line ef being 3.4 /* from the line cd and parallel to it. Then the points 

 x and y are where the surface-line of the conidium cuts the line ef. It is 

 obvious that as the line xy increases in proportion to the length of the conid- 

 ium, gh, the conidium more nearly approaches the form of a cylinder; and 

 as the line xy becomes proportionately shorter the conidium becomes less 



xy 

 like a cylinder. The ratio may therefore be termed the coefficient 



