aquilinuM , with some Notes on Marsihci quadrifoha , drV. 575 
substances of the middle lamella become altered, so that they stain darkly 
with protoplasmic dyes. The portion of the middle lamella which is thus 
altered separates the callus rod into two halves, and is called the median 
nodule. The slime strings may be further bored out until they become 
very thick, and meanwhile the median nodule is apparently dissolved away 
and callus deposited as a tubular lining right across the pit-closing membrane. 
In Wistaria the wall between adjoining slime strings may be completely dis- 
solved away, and several small strings may coalesce to form one large one. 
In Vitis the original slime strings remain independent. Callus is present on 
the Sieve tube side of the pits connecting sieve tubes and bast parenchyma 
cells. The sieve tubes of Vitis function from year to year, and become 
plugged during the winter with callus, which is dissolved away in the following 
spring when the phloem resumes its activity. 
In the light of the foregoing facts, an examination of the sieve tubes of 
Pteridium aquilinum was undertaken in order to determine whether the 
protoplasm of adjacent sieve tubes is, or is not, continuous across the sieve 
plate, and if it is so, what relation such a connexion bears to the dis- 
tribution of the callus. It was also hoped, if possible, to throw some light 
upon the nature of the refringent granules. Pteridium aquilinum was 
chosen for the purpose, because it possesses exceptionally large sieve tubes ; 
for the same reason Marsilia quadrifolia was also selected. In the latter 
case everything seems to indicate that the development of the sieve tubes 
closely resembles that of Pteridium , but the actual sequence of development 
was not worked out. The absence of secondary thickening in Ferns makes 
the accurate seriation of the various developmental stages of the sieve tubes 
a rather tedious business, as a number of blocks have to be cut, and after 
preparation the stages contained in them are often found to be only duplicates 
of those already obtained. The development was worked out in the rhizome, 
as it contains all the older stages as well as the young ones. 
Methods of Research. 
The methods used were, in the main, those introduced by W. Gardiner 1 
and adapted by A. W. Hill 2 and M. G. Thoday . 3 
A dilute solution of iodine in potassium iodide was used for fixation, 
and swelling was effected by treatment, for two or three weeks, with a much 
stronger solution of iodine. Kolossow’s mixture was used for mordanting, 
and for staining, safranin and London blue. Excess of safranin was washed 
out in dilute glycerine, and the washing-out process is continued in the London 
blue . 4 In cases where staining with safranin was insufficient the sections 
1 Gardiner (’97), p. ioo ; (’98), p. 504. 2 Hill (’01, ’08). 
8 Thoday (Sykes) (’08, ’ll). 
4 For the use of this stain as a callus reagent and for washing out safranin from cell- walls, see 
Sykes (’08), p. 299. 
