988 Groom .— The Evolution of the Annual Ring and 
and contrast with the outermost ones of the immediately preceding ring. 
The remaining part of the ring is rather uniform as regards the size of its 
vessels, for larger and smaller ones are distributed apparently indifferently. 
In both species a layer of flattened fibro-tracheides marks the outer 
boundary of the ring. 
In all the species so far described in the three classes the limiting layer 
of flattened tracheides is visible and continuous. But in the next two types 
this layer becomes here and there obscure, because the flattening of the 
fibro-tracheides ceases over certain tangential distances. 
In contrast with all the species so far dealt with—which are American 
— Q. semecarpifolia , Sm. (PI. LXXIV, Fig. 8), is Indian. It is apparently 
sub-evergreen, as Gamble describes it as ‘ evergreen ’, while Brandis states 
that it is 4 sometimes leafless for a few weeks in spring \ The pore-zone in 
the spring wood is indicated by widely separated vessels that are often 
the largest in the ring and nearly always larger than the outermost ones 
of the immediately preceding ring. Yet in some of the radial series of a ring 
the widest vessels lie in the middle zone. The limiting band of flattened 
tracheides is interrupted only for short distances, and even at these places 
the boundary of the ring is recognizable by reason of the shorter length of 
the ray-cells, their thicker terminal walls, and their richer contents. 
Class IV. Evergreen. 
In Class III there does not appear to be any parallelism between the 
longevity of the leaves and the distinctness of the pore-zone, for the tran¬ 
sitional types of wood belong to evergreen and sub-evergreen types. Yet 
the Indian species, which is sub-evergreen, shows more definite tendencies 
towards the production of a pore-zone than do the four truly evergreen 
Indian species about to be described, for these reveal, coupled with the 
absence of a pore-zone, a frequent transference of the maximum diameter of 
vessel to the middle zone of the ring, or show no recognizable relation 
between diameter and position of the vessels in the ring. 
Q. glauca ) Thunb. (PI. LXXV, Fig. 9), and Q. lainellosa , Sm. (Fig. 10), 
have rings that are recognizable by reason of the bounding line of flattened 
tracheides. Frequently the vessels in the outermost part of the ring are dis¬ 
tinctly smaller than their predecessors, but apart from this there appears to 
be no rhythmic succession in the ring as regards calibre of vessel. In both 
species the layer of flattened fibro-tracheides is never very distinct; and 
in places the fibro-tracheides forming the boundary of the ring have not 
the flattened form (as was noted by Abromeit in Q. lamellosa ) ; yet, espe¬ 
cially in Q. glauca , these bounding fibro-tracheides show a greater number 
of bordered pits than do the fibro-tracheides immediately preceding them. 
The remarks made in reference to these two species also apply to 
the rings of Q . incana> Roxb. (Fig. 11), in which, however, the partial 
