1911-1?.] BOTANICAL SOCIETY OF EDINBURGH 63 



tig. 1). except the basal part (nearest the primary axis), 

 which later gives rise to the intercalary meristem in the 

 elongating peduncle of the capitulum. 



How shall we interpret this peculiar development \ Two 

 different views are obviously possible. 



It may be that the lirst leaf (a, rig. 1) arising from the 

 growing-point of the capitulum is the first leaf of the in- 

 volucre of the capitulum, the individual leaves of which, 

 united together, form an apparently entire envelope. This 

 is the concept of Rostowzew. which, however, has thus far 

 to be supplemented, that the " torsion of the capitulum " — 

 if one assumes such a thing — is not a subsequent event. 

 but it existed from the first ; in other words, it is " con- 

 genital." Rostowzew does not mention that if this inter- 

 pretation be assumed, then the lateral capitula of the 

 Ambrosia would present one of the rare ca^es in which 

 the "■ bracts " (Deckblatter) of lateral shoots have aborted 

 completely. This occurrence is known in the flowers of 

 the crucifera?, etc., in which one can correlate the abortion 

 of the bracts with excessive crowding originally under- 

 gone by the flower-primordia before the elongation of the 

 axis of the inflorescence. In the case of Ambrosia, this 

 would be all the more striking because the female in- 

 florescences possess well-developed bracts, although they 

 are quite as closely crowded together as the male ones. 

 In no instance have I observed in the male capitula even 

 a trace of a subtending bract. 



The second possibility is that the first leaf (indicated by 

 a in tig. 1 ) is the bract (Deckblatt) of the capitulum. The 

 manner in which (a) develops from the primordium that 

 gives rise to (a) and (6) is similar to what occurs frequently 

 in flowers and inflorescences. In this instance the bract, 

 so to speak, is late in being formed. It does not develop 

 in advance of its axillary shoot, but from a primordium 

 common to both. The thing that surprises one is that 

 the axillary shoot should arise on the under side of the 

 bract, and not. as usual, on the upper side. But, after all. 

 this is no more wonderful than the " congenital torsion " 

 already referred to, nor is it quite without precedent 

 amongst other Dicotyledons. The remarkable flowers of 

 Erythrochiton hypophyUa/nthus are situated on the under 



