A STUDY OF PLASTIDS AND MITOCHONDRIA 
227 
similar to the young stages of the oil bodies but which differ from them in 
their chemical properties. Since, in addition to the Anlagen of the oil 
bodies and to these structures which are similar to them, Garjeanne men- 
tions also the Anlagen of the chloroplasts, it would seem that he believes 
that there are present in the cells of some liverworts, at least three varieties 
of specifically different granules. 
Rivett ('18), in an account of certain observations upon Alicularia 
scalaris, finds that the results of staining or fixing the entire leaves with 
2 percent osmic acid confirm the view that the oil is secreted in vacuoles." 
This author also finds certain refractive granules present in the cells of both 
the growing point and the older leaves, which differ in their chemical reac- 
tions from the young oil bodies, apparently agreeing with the observations 
of Garjeanne in this respect. In the meristematic regions a "chondriome- 
like structure" was observed, but no evidence was found "that the chon- 
driosomes were either transformed directly into plastids by a secretion 
within their own substance, or that they are the instigators of secretory 
action on the part of the protoplasm." No evidence was found that the 
refractive granules were chondriosomes, ''since their appearance in the 
stained mature cells is quite different from that of the chondriome of the 
actively dividing cells." 
As already noted, mitochondria have also been described in the liver- 
worts by Scherrer ('13) and by Mottier, ('18), neither of whom was able 
to find a^ny connection between these bodies and the chloroplasts. Scherrer 
made a special study of Anthoceros, a form which possesses the greatest 
interest since it has a ''pyrenoid' in some of its chloroplasts, suggesting a 
close relationship with the algae in respect to its method of starch formation. 
The pyrenoid of Anthoceros has been described by McAllister ('14), 
who finds that it consists of from 20 to 300 minute lenticular bodies, which 
lie near the center of the chloroplast and which stain bright red with safranin. 
McAllister states that there can be no doubt that these bodies are trans- 
formed directly into starch, since ''there is a gradual change of the color 
reaction, from the brilliant red of the pyrenoid bodies to the blue of the 
starch grains." On the other hand, he says that, in the cells of the arches- 
porium, in the spore mother cells, and in the assimilative cells of the sporo- 
phyte, starch is formed without the intermediary action of a pyrenoid — 
apparently arising de novo in the chloroplasts. 
The observations of Davis ('99) agree with those of McAllister in this 
respect, since be states that the first clear indication of the chloroplast in 
the spore mother cells of Anthoceros is the sharp staining of the starch 
grains — purple with the gentian violet. 
McAllister states, further, that there is no doubt that if the Anlagen of 
these bodies (the starch grains of the spore mother cells) are present in the 
plastids of the archesporial cells, they are too minute to be distinguished with 
the highest magnifications. This, however, does not necessarily follow, 
