992 Groom— The Evolution of the Annual Ring and 
The indistinctness of the annual ring in certain evergreen oaks, so far 
as it is due to lack of pore-zone, has long been known and is specifically 
mentioned by Sanio ( 5 63 ) in his description of the wood of Q. Ilex. Abro- 
meit (loc. cit.) did not, however, fully realize the close connexion between 
leafing habit and existence of a pore-zone, because he was misled as to the 
habit of certain species. For he states that of the species examined by 
him only in the following evergreen species is the spring-wood marked 
by wider vessels : Q . Wislizeni , Q. castaneaefolia , Q. glandulifera , Q. ser - 
rata , and to some extent Q. cuspidata and Q. agrifolia. This list requires 
correction. On the one hand Q. serrata is not evergreen, and on the other 
hand Q. hypoleuca is evergreen yet possesses a distinct pore-zone ; and 
curiously Abromeit in the special part of his paper mentions both these 
facts correctly. 
Evolution of the Medullary Rays. 
The wood of all species of Quercus has uniseriate, thin, shallow rays, 
but that of only some species possesses in addition multiseriate, thick, high 
rays. It is possible that the ancestral types of Quercus belonged to the 
thick-rayed or to the thin-rayed type. I propose to trace the series of types 
on the momentary assumption that the line of evolution has been from the 
type with broad high rays to that with exclusively narrow uniseriate ones, 
not because I adopt this view, but because the broad-rayed type of Quercus 
is the familiar one, and because in the solitary type that appears to point 
very strongly in one direction, that direction is from the broad-rayed 
to the narrow-rayed type. 
In some types, Q. Macdonaldi (PI. LXXV, Fig. 13) for instance, the 
broad rays are high multiseriate plates of parenchyma, showing, in the 
tangential section, deep vertical expanses of ray-parenchyma devoid of any 
fibres. 
In other types fibres run in the broad ray in various directions. These 
subdivided rays may be ranged into two classes, according to the main 
direction of the fibres :— 
(a) Fibres transverse or transversely oblique , forming in tangential 
section lines or bands that are uniseriate, biseriate, triseriate, up to multi¬ 
seriate. With increased thickness of the bands of dividing fibres a climax 
is reached when in tangential section there is a series of obliquely or 
vertically superposed rays more or less fusiform, so that it becomes difficult 
to decide whether to describe the series as a single divided ray or as 
a number of superposed rays. 
(b) Fibres vertical or vertically oblique, leading finally to the division 
of the broad ray into a series of small rays arranged side by side. 
The broad rays often show transverse, vertical, and intermediate 
dividing fibres, but usually one type of division is more prominent. 
