176 Samuels. — A Pathological Anatomical Study of 
Furthermore, the so-called crystal tubes ( Kristallschlduche ) of Kohl 
are not tubes at all, but are hypertrophied cells which are formed in the 
parenchymatous tissue after the fusion of a number of cells of a certain cell 
complex. They have not the form of tubes only, as Kohl contends, but 
they are round, oval, or curved, according to the position of the cells. 
Having studied the developmental history of the crystal cysts, I have 
decided to make a careful study of other so-called Kristallschlduche 
described by Kohl and others, and shall not be surprised to observe the same 
phenomena. 
According to Dalitsch’s study on the leaf anatomy in the Aroideae, 
the raphids are formed in those cells which lie vertically on each other. 
Such cells are fused together, and the crystals are embedded in a dense mass 
of protoplasm. Dalitsch (p. 345 ) observed this in Philodendro7i longi- 
laminatum. 
In Fig. 13 , Plate III of his paper ( 8 ) the author demonstrates a case 
of several raphid colonies in a parenchyma cell near the epidermis of 
the leaf. 
Recently a similar case has been described by Porsch. Engler (12), who 
mentioned Porsch as the author of the crystal cells in Philodendron Sellowii , 
did not give the title or the year of issue for that particular publication 
(p. 28 ). Probably neither Dalitsch nor Porsch saw any nuclear fusion, and 
did not carefully study the development of the crystal cysts, but, with 
Hansteen, regarded the fused cells as crystal tubes. 
The crystal colonies in A. scandens and A. Scherzerianmn are not 
in a common sac, as with those found in the regular raphid cells just dis¬ 
cussed. Besides this distinction, every needle is separated from the plasma 
by a plasma and not a cellulose wall. 
The solution of the cell wall in pathological cells is of course the cause 
of the formation of a number of crystal colonies in a single crystal cell, and 
probably the cause of their position, because the pressure of the protoplasm 
of the neighbouring protoplasts acts doubtless in different directions, 
whereas the fusions of the protoplasts do not all take place at the same 
time. The result is that here we are concerned principally with abnormal 
regulatory phenomena located in certain cell regions of the tissue. 
On the piercing of the cell wall between two or three contiguous 
parenchyma cells, cell fusion takes place, and the calcium oxalate, together 
with the plasma substance, is forced near to the fusion point of the nuclei, 
the position at which the first crystal colony is to be formed. 
It is highly probable that the union takes place in the immediate 
vicinity of the break in the cell wall or possibly at the opposite side of the 
break in the wall. Immediately after the union it may be that, in the first 
case, the two nuclei are carried along in the direction of the plasma move¬ 
ment, after which they fuse together. The first colony which appears indi- 
