266 SCIENCE PROGRESS 



In the phloem of the Gymnosperms and of the Vascular 

 Cryptogams such sister cells are not found. In the former 

 group of plants the companion-cells are probably represented 

 by certain cells known as the "albuminous cells," on account of 

 the great amount of proteid material in their contents. In the 

 Vascular Cryptogams the phloem contains, beside sieve-tubes, a 

 number of cells rich in proteid and these may correspond to 

 companion-cells. 



Unlike the sieve-tubes the companion-cells do not form 

 continuous strands for any considerable distance and do not 

 present any very obvious features which might be held to indi- 

 cate that they form paths for longitudinal conduction. They are 

 often situated between the sieve-tubes and the parenchyma ot 

 the phloem, the cortex or the medullary rays. This distribution, 

 their dense protoplasmic content, comparatively large nucleus 

 and intimate connection with the sieve-tubes by means of 

 protoplasmic threads, suggest that they may be concerned in 

 negotiating the exchange of materials which goes on between the 

 sieve-tubes and the tissues surrounding them ; but no direct ex- 

 perimental proof of this has yet been brought forward. Follow- 

 ing the sieve-tubes and companion-cells out into the veins of the 

 leaf, it is found that the diameter of each decreases. That of the 

 sieve-tubes, however, does so more rapidly than that of the com- 

 panion-cells, with the result that towards the ultimate endings 

 of the finest veins the companion-cells are considerably larger 

 in cross section than are the sieve-tubes (figs. 9, 10). Finally, 

 sieve-tubes are absent and " transition-cells " occur, which are said 

 by Haberlandt to arise by the failure of the mother cell to 

 divide. Fischer thought that these transition-cells are the seats 

 of albumen formation. As proteids are formed in the leaf and 

 large "nuclei such as the transition-cells possess are frequently 

 associated with great metabolic activity, this theory is quite a 

 plausible one, though actual proof is not available. 



Investigators seem to agree that the companion-cells do not 

 contain starch. In the course of my own observations, whilst 

 I have repeatedly found sugars in sieve-tubes, it has never been 

 possible to determine clearly their presence in companion-cells or 

 albuminous cells, however distinctly visible these may be. The 

 exact functions of these cells are by no means clearly understood. 



Relation of Phloem to other Tissues. — As a rule the amount 

 of phloem present (though not necessarily the number of sieve- 



