140 
the frequent recurrence of these variant sectors commanded attention, trans- 
fers were made from several of them to freshly poured agar-plates, and a 
transfer from the normal portion of the colony was added to each of these 
plates at a distance of about 2 cm. from the other transfer. The variant 
transfer was then marked on the bottom of the doubly occupied plate as 
M (indicating mutant), and the normal transfer as O (indicating original). 
In all the early transfers the M transfer resulted in a colony (M1) of decid- 
edly slower growth and more profuse conidial production than that pro- 
duced by the O transfer. The two colonies also differed markedly in general 
appearance owing to minute single differences which were often difficult to 
analyze, but which in the aggregate constituted distinctions which were so 
well-marked and obvious that at first sight one would say that the two 
colonies were those of two distinct fungi (Pl. X XIII, lower fig.; Pl. XX VII). 
When these colonies grew to fill, or nearly to fill, the plate, transfers from 
them were made to new agar plates, and later, transfers from these second 
plates, and so on, the series of transfers being a long one. It was found 
that the differences appearing in the M1 and O1 colonies were usually 
maintained on succeeding plates. These findings led to the tentative 
assumption that forms in the variant sectors were mutants or saltants of 
a more or less permanent nature, and a more serious study of this phe- 
nomenon was undertaken. 
In their origin the variants or saltants always appear as sectors which 
differ from the portion of the parent colony adjacent to them (see Pl. XXII 
—XXV). To the naked eye the most common deviations from the orig- 
inal type are in density, color, and rate of growth. Closer observation, with 
the microscope, frequently shows variation in the grouping, size, and shape 
of the conidia, and in the branching of the mycelium. Quite often many 
small sectors of divergent character appear at the edge of a large colony, 
especially on a plate that is beginning to dry. Many of these divergencies 
are merely modifications due to local environmental changes, and whether 
they are more can be determined only by close study of their behavior in 
subsequent transfer or transfers. Closer consideration of the characters 
involved in these saltations is best deferred to the following topic. 
In following this discussion it must be borne in mind that M refers to 
the variant sector on the plate on which it originated; M1, to the colony re- 
sulting from the first transfer from M; M-2, to that resulting from the first 
transfer from M1, and so on; and that O refers to the original colony in 
which the M arose. It is my custom to give the saltant a serial number 
(writing this on the plate in which it was found), and, usually, to transfer 
