Nov., 1922] LEVINE LAMELLAE IN AGARICUS CAMPESTRIS 523 
figure 76, we have three groups of paHsaded hyphal cells and the gill cavities 
have entirely disappeared; that is, we have reached that portion of the 
hymenial primordium in which the palisaded cells while oriented have not 
as yet sufficiently separated to form a cavity. There is, of course, con- 
siderable fluctuating variability in the arrangement and the size of the gill 
chambers, and the sections are by no means exactly parallel to the longi- 
tudinal axis of the young carpophore. This axis, again, need not be a 
perfectly straight line. These facts will account for many apparent incon- 
sistencies in the sections. It is, however, sufficiently clear from the figures 
that these rudimentary gill cavities arise at the region where the tips of 
the oriented palisaded cells meet. They become larger by an increase in 
the number of the palisade hyphae which seem to take their origin in the 
pileus and the tramal cells. The result is an arched rudimentary hymenial 
surface, under which the gill cavity becomes larger and larger as this hy- 
menial surface increases and the walls of the two adjacent arches with their 
common trama become a single gill. In perspective we should have a 
series of these radially arranged arches and tramal plates, older and larger 
near the stipe and younger and smaller toward the margin. The tramal 
plates remain firmly attached to the stipe tissue below since their hyphae 
run through from pileus to stipe. The growth of the pileus rudiment is 
centrifugal, so that, while the rudimentary arch or gill cavity may be a 
wide opening near the stipe, it is just unfolding or opening at the margin 
of the pileus. 
Tangential sections of older buttons are represented by figures 77, 78, 
and 79. In the middle of such a section the gill chambers are distinct 
openings (figs. 78, 79), taking the shape of Roman arches; toward the 
right and left they dwindle away, and we have only a pocket of hyphae with 
their tips pointing to a common center and no visible opening (fig. 77), 
In the last mentioned figure we find the two gill cavities still unopened or 
in a very rudimentary state. Each palisade cell contains two nuclei which 
stain well with Flemming's triple stain. There is no cytological indication 
at this stage that the older cells are to be found near the bases of the arches 
or rudimentary lamellae, as Buller (1909) suggested for the mature gills, 
and yet this observation does not indicate an inconsistency, for it is quite 
possible that the hymenium may mature and develop later from the margin 
of the pileus and outer edge of the gill upward, and be independent of the 
time of origin. 
As we proceed still farther toward the center of the plant a large number 
of gill cavities appear, as shown in figure 79, which represents the maximum 
number that appear in a longitudinal tangential section of this button. 
Figure 77 becomes more intelligible when studied in connection with 
diagram 3, which represents a top view of the carpophore in the earliest 
stages of development. The outermost space (a) between the two outer 
circles represents the region of the fundamental and undifferentiated tissue. 
