Cliondriosomes (Mitochondria)and their Significance 105 
was given when the merest traces of chlorophyll appeared, for 
instance, when etiolated seedlings were exposed to light for a few 
hours. Using both the Meves’s staining method and the silver 
method, he traced the formation of chloroplasts from cliondriosomes 
in the young leaves of a considerable number of flowering plants, 
and fully confirmed Lewitsky’s results. 
Forenbacher (1911) studied the origin of chloroplasts and 
leucoplasts in the stem and root of Tradescantia , for fixing which 
he used chiefly a modified form of Benda’s fluid consisting of 
chromic acid and osmic acid, acetic acid being omitted, but also 
used absolute alcohol which he found to give good results, while in 
addition to Meves’s modification of Heidenhain iron-haematoxylin 
he used the well-known triple stain (safranin, gentian violet, orange). 
He began with fully developed plastids and proceeded towards the 
growing meristem, and found every stage between these and the 
cliondriosomes of the young growing cells, the intermediate forms 
being dumb-bell structures exactly like those described by Lewitsky. 
Lewitsky published two further papers in 1912. In the first 
of these he gives side by side photographs of living cells of Elodea 
and of cells fixed and stained by the Benda-Meves method, to show 
that the chondriosomes are not in any sense artefacts but can be 
seen quite clearly in suitable living cells. He then tried various 
fixative fluids and divides them into two groups—those which give 
the real structure of the cytoplasm and show the chondriosomes in 
their natural condition (Benda’s fluid, the same minus acetic acid, 
Altmann’s fluid, 0 5% osmic acid, 10% formalin, weak Flemming) 
and those which give various deformed products and artefacts and 
are therefore unsuitable for work on chondriosomes (absolute 
alcohol, Carnoy’s acetic alcohol, mixtures containing corrosive 
sublimate, silver nitrate plus pyrogallic acid, strong Flemming). 
These various fluids were tried both with microtome sections and 
with living cells (the axillary scales of Elodea were used) examined 
directly under the microscope. In the second paper Lewitsky 
points out that many of the methods hitherto used in the study of 
the development of plastids are quite unsuitable, since they cause 
deformation of the chondriosomes from which the plastids originate. 
As in his earlier work he found that the plastids arise as swellings 
of the ends of filamentous or rodlike chondriosomes, and he gives 
photographs of the young choroplasts in the “ chondriokont stage.” 
During 1911 and 1912 Guillermond published a series of papers 
giving the results of an elaborate investigation of chondriosomes 
with special reference to their function as the forerunners of 
plastids, and confirmed and greatly extended the work of Lewitsky 
and Pensa. He used various methods, but chiefly that devised by 
Regaud for animal chondriosomes. He examined plants belonging 
to all the main divisions of the vegetable kingdom and found 
chondriosomes in nearly every case (reference will be made later 
to his work on chondriosomes in fungi), but his first papers dealt 
mainly with the behaviour of these bodies in germinating seeds, 
particularly those of Gramineae. Here he found that the chondrio¬ 
somes in the cotyledon mostly remained unchanged, while those 
in the meristem cells mostly disappeared as plastids were formed 
at their expense. In the cells at the base of the young leaf these 
bodies at first lie close to the nucleus as short filaments which 
during mitosis arrange themselves around the spindle, but in 
