28 Hardy, Disiribuiion of Leaf Glands in Acacias, [vor'xxtx. 



an appropriate link, the l)ipinnate juvenile foliage of the 

 young plants of the latter. Species such as A. mclanoxylon, 

 A. implexa, A. pycnantha, and many others have ordinary 

 bipinnate foliage when young, and,, occasionally, reversionary 

 foliage of this character may be produced on a branch of an 

 adult tree bearing, normally, phyllodia only, as the result of 

 injury — a local "second childhood"! As we know the 

 phyllode to be a modification of the vertically expanded 

 petiole and rhachis, or petiole only, we look for the gland on 

 the upper edge of the expansion or ala of the transition leaf. 

 In all cases examined by me the gland was situated at or near 

 the base, and none was found along the rhachis, from which 

 it might appear that, in the reduction of the bipinnate leaf 

 to the bare phyllode, there is, in the evolutionary })rocess, a 

 precedent loss of the rhachial glands. 



Phyllodia of Mature Plants. — Next we may turn to the 

 phyllodia of mature plants and find that the position of the 

 gland varies, but with a fair amount of constancy in certain 

 species, while in others the variability is more marked. In 

 A. longifolia, var. sophorce, a count of many hundreds showed 

 3 per cent, nearer |-inch fi'om the base than the prevailing 

 J-inch. Generally speaking, the species with long phyllodes — 

 e.g., spp. Maideni, stenophylla, linearis, cyanophylla. saligna — 

 have the gland at or near the base, while the cases in which 

 the single gland is further removed may be observed in phyllodes 

 having, in proportion to length, a greater lireadth — e.g., spp. 

 podalyrcefoUa (with almost broad-elliptical i)hyllodes), myrti- 

 folia, and A. cultriformis and A. prarissima (with small, broad, 

 angular or sub-angular, oblique phyllodes). but. like many 

 other rules, this has its exceptions. 



Certain species show uncertainty as to both number and 

 position of the glands. Thus, in A. amana (fig. 7) there may be 

 only one, more often two or three and sometimes four, rarely live. 

 In this species, when only one gland is present, it is, probably 

 without exception, nearer the base than the apex. If two glands 

 are borne they may be both below the middle, or may be roughly 

 equidistant from base and apex rcs])ectively — e.g., A. decora 

 (fig. 10) — while three may divide the phyllode margin into four 

 about-equal parts. 



(ilands do not appear doubled on the phyllode margin so 

 Ireciuently (if at all) as on the rhachial ridge (fig. iq) of the 

 compound leaf. 



It is improbable that in Australian acacias more than fi\'e 

 glands are to be found 01. one phyllode, though five have l)een 

 noted on specimens of A. ama-na by the present writer. This 

 brings us to a case of variation, in which the gland may 1)c 

 close to the base and so subordinated as (o he almost lost to 



