28 o 
Journal of A gricultural Research 
Vol. II, No. 4 
II. Terminal chlamydospores absent, conidia ochreous to salmon colored. 
a. Conidia with curved apical end constricted like a flask neck, heel of the pedicel¬ 
late base not prominent. Section Discolor. 
1. Intercalated chlamydospores rare and not in clusters. No carmine color on 
sterilized potato tuber.. (PI. XVI, fig. L) Fusarium incarnatum (Rob.) Sacc. 
2. Intercalated chlamydospores occur singly and in cluster chains. A carmine 
color develops on sterilized potato tuber. 
(PI. XVI, fig. J) Fusarium culmorum (W. G. Sm.) Sacc. 
b. Conidia with a prolongated and pointed apical end and with the heel of the 
pedicellate base prominent. 
x. Parabolic to hyperbolic curves prevail in conidia seen in side view. No 
carmine color on steamed potato tuber. Section Gibbosum. 
a . Apical end of conidia curved, subfiliform. 
(PI. XVI, fig. P) Fusarium caudaium, var. volutum, n. var. 
b . Apical end of conidia curved, filiform. 
(PI. XVI, fig. M) Fusarium caudaium, n. sp. 
III. Terminal and intercalated chlamydospores absent. Conidia resemble those in 
the section Gibbosum, but hyperbolic curves are seldom pronounced. A car¬ 
mine mycelium color on steamed potato tuber. Section Roseum. 
(PI.XVI, fig. G ) Fusarium acuminatum Ell. and Ev. 
B. SPECIES OF FUSARIUM WITH KNOWN PERFECT FORM 
1. Conidial stage similar to the section Martiella of the genus Fusarium, but with 
a subpedicellate base. Section Pseudomartiella of the genus Hypomyces. 
a. Conidia largely 3-septate. Perithecia averaging in size 350 to 450 by 275 to 
375/1. Ascospores, 10 to 15 by 5 to 6.75/1. 
(PI. XIII, fig. J) Hypomyces cancri (Rutg.) n. comb. 
b, Conidia largely 5-septate. Perithecia, 225 to 375 by 175 to 300/1. Asco¬ 
spores, 10 to 13 by 4.5 to 6/i. 
(PI. XVI, fig. H) Hypomyces ipomoeae, (Hals.) Wollenw. 
2. Conidial stage similar to the section Discolor of the genus Fusarium, but 
chlamydospores absent. Genus Gibberella. 
(PI. XVI, fig. 0 ) Gibberella Saubinetii (Mont.) Sacc. 
This key might have been based entirely upon the morphological char¬ 
acters and curvature of the conidia, but since the color reactions offer a 
simpler, though less trustworthy means of identification, they have been 
employed. The key, therefore, should be regarded only as an aid in 
identification, not as a guide to the morphology, which has been dis¬ 
cussed in the diagnosis and illustrated in detail in the illustrations. 
In Table II the average size of the various spore types has been given 
approximately to allow a survey of the differences between the species. 
The spore diameter is an important factor for the determination of species 
of Fusarium, as may be seen in comparing equiseptate conidia of Fusa¬ 
rium culmorum and F . camptometachroum. The conidial length is less 
significant. The measurements of chlamydospores recorded are confined 
to their cross diameter, the two dimensions not varying much in the 
almost spherical unicellular spores. Ovoid and 2-celled spores have a 
major axis which consequently has a higher average length than the 
diameter of spherical spores shows in Table II'. 
