1901] MICROSCOPICAL JOURNAL. 289 
then cut just above and just below one of the plates. But 
it is better to remove the parts outside of a a and dD b of 
fig. 1 and dissect out one of the diaphragms to examine 
by itself, 
By suitable focussing it will be seen that the cells are 
somewhat convex. They not unfrequently contain dis- 
tinct crystals of calcium oxalate as shown at 4g, fig. 3; and 
occasionally cells filled with raphides may be seen lying 
on the plates, as at e. 
But where the Musa is not available, good specimens 
of the diaphragms may be readily obtained from the 
Pontederia cordata which grows in shallow waters every- 
where. The petioles which spring from the rhizome con- 
sist mainly of sixty or more air-ways with cross plates 
less than two mm. apart. The flower-bearing stem of 
this plant has, besides about 150 small air-passages, a 
large central one in which the ethmoids are about five 
mm. from each other. After splitting the stem so as to 
expose this central cavity, one of the plates may be cut 
loose around the edges and be removed for examination. 
They are like fig. 3. In almost every transverse section 
of the stem,ethmoids may be seen in some of the smaller 
air-ways. And the same is true of the petioles. 
Fig. 6 shows a cross section of a Pontederia stem of 
three halves the real size. Fig. 7 gives a short piece in 
longitudinal section on the same scale. Fig. 8 repre- 
sents one-sixth ofa cross section of ten times the natural 
size. 
Another of the Pontederiacex, the very prolific Hich- 
hornia crassipes (or speciosa), which has taken possession 
of our Southern bayous, abounds in air-passages with 
ethmoid plates. Inone petiole, 380 air-ways were count- 
ed. The flower bearing stem, like that of the Pontederia 
has a large central passage in which the partitions are 
from 6 to 30 mm. apart. Around this central cavity the 
smaller air-ways are quite as numerous as in the Ponte- 
