136 ©. BOTANICAL GAZETTE [aucust 
Of course, the objection may be raised that the membranes 
were ruptured by the process of injection. This seems to be highly 
improbable, however, since in a special series of experiments, 
devised for the purpose, the writer was unable to rupture the 
membranes, even under a pressure of 250 lbs. per square inch. 
In order to obviate this difficulty entirely, the India ink was much 
diluted with distilled water, and allowed to percolate through 
sticks of freshly cut sap wood‘ under slight hydrostatic pressure 
(0.25 oz.). Under these conditions the India ink’ passed through 
the bordered pits easily and rapidly. 
By experimenting with various methods of cutting, staining, and 
mounting sections, the writer finally succeeded in making visible 
and photographing openings in the membranes (fig. 6) of the pits 
of Larix laricina. The reason that the perforations have escaped 
observation is undoubtedly due to the fact that the pit membranes 
are extremely minute and attenuated, and their structure is ob- 
scured by the thick overlying walls. Unless the embossed portions 
of the secondary walls are cut away, leaving the membranes 
exposed and uninjured, a somewhat difficult undertaking, or a 
necessary combination of light refraction phenomena is secured, 
the perforations remain invisible. 
It should be noted in passing that a perforated structure of 
the pit membranes makes possible an entirely different inter- 
pretation of the penetration of gelatine, paraffin, cacao butter, 
and similar substances into the xylem of conifers. In discussing 
SacuHs’ imbibition hypothesis and the mobility of water in tracheid 
walls, the significance of ELFvinc’s experiments with cacao butter 
and those of ERRERA and STRASBURGER with gelatine has been 
questioned upon the assumption that, because cacao butter and 
‘Tests thade with Pinus Strobus L., P. rigida Mill., Picea rubra (DuRoi) Dietr., 
Larix laricina (DuRoi) Koch. io Saleidbansy (L.) Car Juniperus, virginiana L., 
and Taxus cus pidata Sieb. and Zac 
5 The various commercial India inks vary considerably in the size of sei! carbon 
particles of which they are composed. Although the writer has succeeded in 
penetration with various Asiatic inks, they are usually composed of too coarse sorties: 
and tend to clog quickly, Higgins’ India ink contains much smaller and more uniform 
carbon particles and gives much easier and rapid penetration. In using India inks 
as injection masses the writer has been careful to make sure that that portion of the 
ink which passed through the membranes contained visible particles of carbon. 
