1958] 



THE BOTANY OF THE QUAY AN A HIGHLAND PART III 



63 



The aboYe distinctions were essentially those used by Berg in contrasting 

 these genera, which are otherwise much alike in vegetative characters and in 

 characters of flowers and fruit. It has recently been emphasized anew, however, 

 that certain distinctive characters reside in the inflorescence. 20 In the subtribe 

 Myrciinae the inflorescence is usually a compound or decompound panicle with 

 more or less elongate lower branches and the flowers mostly appearing to be 

 clustered in groups of three near the tips of the panicle-branches. In the genus 

 Myrcia the principal axis of the panicle is as long as or longer than the primary 

 lateral branches, and the panicle as a whole is about as long as wide. In Mar- 

 lierea and Calyptranthes, on the other hand, in a majority of the known species, 

 the principal axis of the panicle shows a strong tendency to abort immediately 

 above the lowest node. The inflorescence, as a result of this, consists of a very 

 short central axis terminated by an abortive bud and with a pair of nearly 

 equal, and elongate flowering branches arising from the node just below the 

 terminal bud. These paired "panicles" in Calyptranthes and Marlierea are, of 

 course, the homologs of the lowest panicle-branches in Myrcia. 



In Myrcia the inflorescence is usually axillary (sometimes appearing terminal 

 because the panicles surpass the leaves at the uppermost node) ; no morphologi- 

 cal distinctions have been observed between axillary and terminal panicles. In 

 Marlierea and Calyptranthes, however, the inflorescence may be (and usually is) 

 axillary, or may be terminal on the same plant. It has not been possible to verify 

 the following observation for all species of any of the genera in question, but it 

 now seems that the occurrence of a terminal panicle is rare in Myrcia and rather 

 common in the other two genera. The terminal panicle (ax contrasted to the 

 falsely terminal condition often seen in Myrcia) is formed as a prolongation of 

 a shoot which is leafy below and not by development of an axillary bud at the 

 terminal node of the shoot. In Marlierea and Calyptranthes the terminal panicles 

 are usually indistinguishable from those of Myrcia, whereas the axillary panicles 

 more often tend to exhibit the paired condition resulting from abortion of the 

 axis. In Calyptranthes at least, the tendency for the abortion of the axis may 

 be carried over to the vegetative branches, with suppression of the central axes 

 and continued groAvth of the axillary branches, so that the genus may often 

 be recognized from sterile material by the repeated forking of the branchlets. 



A further distinction between Myrcia on the one hand, and Marlierea and 

 Calyptranthes on the other, may be found in the distinctive pubescence of the 

 latter two genera. In these (as also in some species of Eugenia) the pubescence 

 in many species is at least partly of two-branched hairs which usually lie closely 

 appressed to the surface and are often attached very near to one end so that 

 their two-branched condition is apparent only upon close inspection. Hairs of 

 this kind occur frequently in Calyptranthes and often in Marlierea, but I have 

 not observed them in Myrcia. 



The principal genera in the Myrciinae are thus delimited by at least three 

 sets of characters which indicate that Myrcia is amply distinct from Marlierea 

 and Calyptranthes; these latter genera, on the basis of the same characters, are 

 more closely related. From study of recently discovered species it is now appar- 

 ent that there is hardly any real distinction between Marlierea and Calyptranthes, 

 so that the line between them must often be drawn rather arbitrarily- it is 



20 Fieldiana Bot. 29: 152-161. 1956. 



