SPIRAL TRACHEIDS 37 



the observer. Usually more than one spiral is developed in the 

 same tracheid, but it does not follow that the full number will 

 be present at all points throughout the length of the tracheid. As 

 between different genera and species, the number usually varies 

 from one to four. This is much less than in the Angiosperms, 

 where they may be as many as sixteen to twenty. Localization 

 also occurs in such manner that the spirals often run in series, 

 these latter being separated from one another by wider intervals, 

 and as this relation is subject to somewhat wide variation within 

 the same species, it follows that it cannot be utilized success- 

 fully for diagnostic purposes, although it is quite possible to rec- 

 ognize and define and utilize the more general differences of 

 distribution in the terms few and distant, numerous and approxi- 

 mate. The variation in distribution above referred to is largely 

 expressed in the fact that in the tracheids first formed in a sea- 

 son's growth the spirals will always be most widely separated, 

 while those which are formed later constitute a more compact 

 series. This fact becomes prominent wherever such structures 

 can be observed through a. considerable radial extent of wood, and 

 it is therefore particularly well shown in growth rings, though it 

 may also be seen in the protoxylem region when the latter is of 

 great radial extent, as in Cordaites. Thus in Taxus or Torreya 

 or Pseudotsuga it may be seen that in passing from the earli- 

 est spring tracheids to the last-formed summer wood there is a 

 graduated condensation of the spirals which agrees with the 

 relative rate of development in the tracheids themselves. A 

 similar variation is to be seen within the limits of an individual 

 tracheid in such way that the spirals in the central region are 

 more widely separated than those nearest the extremities. This 

 has been adequately explained by De Bary (13, 157), who has 

 shown the more distant coils to result from stretching of the 

 walls during the period of very active development, but subse- 

 quently to the formation of the spirals ; while the condensed 

 forms would be an expression of a slower rate of growth in the 

 cells or in special regions of them, in consequence of which the 

 spirals more nearly retain their original relations to one another. 



