ANATOMY OF THE GENUS SALICORNIA. 329 
collected by him *, we were able to investigate the morphology and anatomy 
of the vegetative structures of one species of the remaining genus 
flalocnemum. 
Halocnemum  strobilaceum, Bieb., shows the feature, unique among the 
Salieorniez, of the differentiation of long and short shoots. The long 
shoots of the plant (Pl. 15. fig. 5) were covered with small, fleshy rosette- 
like structures, which at first sight strongly resemble flower buds. These 
rosettes are decussately arranged upon the long shoot, each consists of four 
or five pairs of opposite fleshy leaves, each pair alternating with the pair 
above (Pl. 15. fig. 5). The leaves in a rosette show no trace of lateral 
fusion, and adnation is impossible owing to the extreme reduction of the 
internodes in the crowded bud, for the rosettes are merely leaf buds in which 
the leaves are succulent; the short shoots originate by the further develop- 
ment of these buds (ef. Pl. 15. fig. 6). The young short shoot is made up of 
jointed segments very similar to those of a Salicornian shoot +, each segment 
of the axis having two free tips, a free sheath, and an adnate foliar sheath 
justas in the genus Salicornia. The transition between the free leaves of 
the rosette which remain at the base of the shoot, and the fused adnate leaves 
of the segments, takes place gradually, and the two appear to be undoubtedly 
morphologically identical. The “ succulent cortex " in the very closely 
allied genus Salicornia is strictly homologous with that of Z/aleenemwm 
strobilaceum. In the latter species (as is shown later on in the paper) it is 
undoubtedly foliar in origin, so that it must be considered as of morphologi- 
"ally foliar nature in the former genus also. In the older short shoots each 
leaf bears in its axil a bud, the two buds of the pair of leaves extending 
round the whole circumference of the stem (Pl. 15. fig. 7). The close agree- 
ment between these axillary buds and the rosettes of the long shoots, both as 
regards arrangement and anatomy, leaves no doubt that the latter represent 
the former in an older stage of development, thus the “long shoots” with 
their decussately arranged rosettes represent the older “ short shoots " in which 
the leaves have fallen, leaving only the axillary buds to clothe the stem. 
Thus it would appear that for developmental reasons, on anatomical 
grounds, and in comparison with nearly allied genera, the “succulent 
cortex ” of the genus Salicornia must be considered as foliar in origin. 
It may be well to state in this connection that although we believe that 
the bulk of the Salicornian shoot is foliar in origin, yet we do not consider 
that this fact should be used as evidence in favour of the * phyton theory " 
which has been used to explain the origin of the leaf of higher plants. We 
regard the Salicornian condition asa specialisation or a secondary adaptation 
ж At the mouth of a shallow wadi opening on to the Gulf of Suez, in April 1901. 
+ The resemblance to S. glauca is very close. 
