36 ROYAL SOCIETY OF CANADA 



mined in the two ways and see how far the results may be reconciled 

 with one another. 



Tlie anatomical details of the vascular cylinder wliich may be 

 held to possess value for diagnostic purposes, stated in the supposed 

 order of development, are as follows: — 



1. Spiral tracheids. 



2. Uniseriate rays. 



3. Resin cells. 



4. Ray tracheids. 



5. Resin passages. 



6. Fusiform rays. 



7. Thyloses in the resin canals. 



8. Cells of the medullary rays of two kinds. 



Of suiordinate diagnostic value. 



9. Resin cells in bands. 



10. Resin cells scattering throughout the growth ring. 



11. Resin cells few and scattering on the outer face of the summer wood. 



Spiral Tracheids. — The spiral bands developed on the inner face 

 of the tracheid wall may be held as primarily designed to afford a 

 greater measure of mechanical support to the structure in which they 

 occnr. In the progressive alteration of the wall, such thickenings 

 tend to disappear, being absorbed in the more general thickening of 

 the secondary growth, which becomes a prominent feature in the 

 tracheids of the Gymnosperms and the great majority of the Angio- 

 sperms. Their obliteration as distinctive markings is, therefore, in 

 direct relation to the higher organization of plants, and as elsewhere 

 shown, their occurrence always serves to establish a definite connection 

 with the lower Gymnosperms or even with more primitive forms among 

 the Cycadofilices.^ Their permanent survival in the Taxaceœ, there- 

 fore, not only serves to give this group a well defined connection with 

 more primitive types, but it also serves to separate it 'definitely from 

 the Coniferge in which, as a whole, such structures are wanting or at 

 least rare, and to which it is inferior in position. 



So far as the North American Conif eras can supply evidence — 

 and the same will hold true also of the exotic Cedrus, Agathis and 

 Araucaria — progressive development has resulted in the final obliter- 

 ation of the spirals and their replacement by bordered pits which arise 

 in natural succession.- It is, however, true that the tendency toward 

 the development of spirals persists in the most highly developed types 

 of the Coniferae. This is expressed in the perlect formation of spirals 



'■ Trans. R. Soc. Can.. II., iv., 39; Ibid, VI., iv.. 57. 



' Ibid., II., iv., 39; Vl., iv., 57; M. OVtic. Jnl., 1869, 67-70. 



