PROTOPLASM THROUGH THE WALLS OF VEGETABLE CELLS. 
845 
successful observation of the threads crossing a thin pit membrane is a matter of 
extreme difficulty; and it also serves to show that in other cases where the pits are 
very small and the membranes very thin this difficulty is so increased as to become 
almost an impossibility. 
But apart from any consideration of the pit membrane, the ease or difficulty of 
observation also appears to depend greatly upon the peculiar characteristics of the 
seed itself. For instance, as I stated at the outset, Plicenix dactylifera and Euterpe 
oleracea are inclined to be unfavourable material. In Euterpe edulis, on the other 
hand, the connecting threads can be easily demonstrated. Of numerous other 
instances, Geonoma, Plectocomia, Areca triandra, Areca catechu, and Cocos nucifera, 
alford examples of cases where difficulties of observation occur. 
Among the most favourable material for examination are the endosperms of 
Bentinckia, Stevensonia, Thrinax, Ileterospatlie, Syagrus, Coryplia, Howea, and 
Lodoicea (see figs. 16, 17, 18, 22, and 25). The degree of development of the middle 
lamella varies greatly. As a rule, in thickened endosperms it attains but little 
development, or, even if this be not the case, it stains but little, and its refractive 
index varies only slightly from that of the general cell-wall. In Calamus, Sabal, 
JRaphia, and Ptychosperma, the lamella is decidedly pronounced. In such cases as 
Stevensonia and Calamus (Plate 69, fig. 24) both the middle lamella and the threads 
are well developed ; and though in them there is some difficulty in determining whether 
the middle lamella is actually perforated by the protoplasmic threads, yet, as a rule, 
careful examination and preparation will decide that in the vast majority of cases it 
can be seen that such perforation does occur, and such examples as Pleterospathe, 
Kentia, Mauritia, or Bentinckia do away with all possible doubt (figs. 16, 19, 22, 
23). As to the manner in which the communication between the endosperm cells is 
established, experiment shows that there are two possible ways which essentially 
depend upon the configuration of the cell. 
In such exceptional cases as Strychnos, Tamus (Plate 70, fig. 33), and Dioscorea, 
where the walls are extremely thick, and, at the same time, devoid of pits, the com¬ 
municating protoplasmic threads run through the cell-wall. A section of such an 
endosperm exhibits the threads, which are seen freely perforating the wall, except at 
the corners of the cell, at the point where the junction of several cells occurs. 
The usual mode of union, however, is by means of pits >jv As I have mentioned 
elsewhere, the presence of pits in the cell-wall, due to unequal thickening, is of almost 
universal occurrence, and it is through the closing membranes of such pits that the 
protoplasmic threads run. This, in fact, appears to be by far the most common and 
typical way in which the continuity of the protoplasm of adjacent cells is brought 
about. 
In other, and perhaps less frequent cases, examples of both modes of connexion 
occur; the communication taking place not only through the pits, but through the 
MDCCCLXXXIII. 5 Q 
