38 Transactions of the South African Philosophical Society. 



is continued by the shoot on the left side. Examining after this 

 the shoot terminating in flower III, we find that the shoot con- 

 tinuing the sympodium is to our right, and the next one is to the 

 left of its relative main axis, &c. By glancing at the sketch of 

 an inflorescence (Plate II, D) and at the diagram which we have 

 just considered, it will be seen that all flowers on the sympodium 

 occupy one side of it (the upper side) but are arranged in a zigzag 

 line, separated horizontally by an angle of 90°, while all ^ leaves 

 occupy the lower side and are arranged similarly as the flowers. 

 While we can thus trace a cicinuus with a sympodial axis composed 

 of the shoots of (3 leaves starting from /3 of the first flower, we must 

 not lose sight of the fact that, as in M. cordifolium, « also has an 

 axillary shoot, which grows according to the same rules as the /3 

 shoot, and that thus the total inflorescence is again a dichasium, but 

 as its axis cannot place itself into the prolongation of the parent axis 

 it appears in its true colour as a lateral branch of the latter, but also 

 developes according to the same laws as the axis on which it is 

 seated. However, it will be seen that the branching from this shoot 

 is much poorer than from /3, and thus here also the inflorescence is 

 a " dichasium mit wickeltendenz." I said before that between each 

 normal shoot and its bract an accessory shoot may arise, and I have 

 to add that, when it developes properly, its growth proceeds also on 

 the same lines as the normal shoots. 



It is a well-known fact that, as a rule, in dichasia corresponding 

 shoots are antidromous. I know no clearer example illustrating 

 this rule than M. angulatum. We will consider the shoots in the 

 axils of a and j3 (Plate II, A). In the axil of the former we have a 

 shoot bearing two leaves, a ' and /3', and the flower II a . As the leaves 

 are opposite one another we cannot decide at first whether we have 

 to connect them by a right or left spiral, but if we look at the 

 sepals we see that these are arranged in a right spiral.* W 7 e also 

 notice that if we try to pass from a ' to /3' and then to the first sepal 

 by the shortest way, we have to turn in the same direction in which 

 the sepals are arranged — we have to traverse a right spiral. On the 

 other hand, we find that the leaves on the shoot in the axil of /3 can 

 be seen to be arranged in a left spiral if we include the sepals of the 

 flower II/3. By looking at the sepals of flowers I and II/3 we see 

 that they are also antidromous, the sepals of I being arranged in a 

 right spiral, those of 11^ in a left spiral, and by comparing succeed- 

 ing flowers on the sympodium we find that succeeding flowers of the 



* A spiral which turns in the same direction as the hands of a watch is called 

 by botanists a right spiral. A left spiral of course turns in the opposite direc- 

 tion. 



