Vol. 10. Xo. 1 MEMOIRS OF THE NEW VORK BOTANICAL GARDEN July 1, 1958 



ANATOMY OF GUAYAJNA MUTISIEAE. PART II. 



Sherwix Carlquist 



In a preceding- paper (Carlquist 1957a), the variation in pollen morphology 

 and floral venation of certain genera of Mutisieae from the Guayana Highland 

 of Venezuela was surveyed. These genera were also incorporated in a study of 

 wood anatomy in this tribe (Carlquist 1957b). In the present study, differences 

 among- these taxa in stem anatomy (other than secondary xylem), node, leaf, 

 and involucre anatomy are described and related to previous data in order to 

 demonstrate the nature of differentiation among these genera, which appear to 

 form a phylogenetic unit (Maguire 1956). In addition, the anatomy of the 

 Guayana Mutisieae is significant in relation to the problem of primitiveness in 

 the Compositae, since it appears that Stenopadus, Gongylolepis, and their allies 

 contain a constellation of characters best interpreted as primitive in the family 

 as a whole. 



ACKNOWLEDGMENTS 



Sincerest appreciation is expressed to Dr Bassett Maguire, who furnished 

 the materials for this investigation from collections of The New York Botanical 

 Garden which were chiefly made by him and his associates on expeditions to 

 Venezuela (Maguire et al. 1953). Certain specimens from other herbaria proved 

 helpful for this study, and thanks are due to curators of the Gray Herbarium, 

 Harvard University, and the Herbarium of the University of California, Berke- 

 ley. These specimens are cited as (GH) and (UC) respectively. 



MATERIALS AND METHODS 



Liquid-preserved materials were available 1 for several taxa (Chimantaea 

 rupicola, Steyermark <& Wurdack 748; Gongylolepis huachamacari subsp. nebli- 

 nensiSj Maguire 37353; Neblinaea promontorium, Maguire. Wurdack & Bunting 

 37016). These collections proved invaluable in showing the nature of cellular 

 contents for vegetative structures. In all other taxa, herbarium material was 

 used exclusively. Good results were obtainable with the following techniques. 

 Stems, leaves, and involucres were cleared in a warm 2.5 per cent NaOH solu- 

 tion until most contents had been removed; subsequent treatment with an aque- 

 ous 25 per cent chloral hydrate solution proved helpful in further clearing and 

 restoring natural shape and proportions to cells. For both liquid-preserved and 

 "revived" herbarium materials, treatment with commercial-strength hydro- 

 fluoric acid was found desirable in most instances on account of the excessive 

 sclerification of many structures. Treatment varied from a few days to several 

 weeks, depending on the structure. After thorough washing, materials were 

 embedded in paraffin according to the tertiary butyl alcohol series of Johansen 

 (1940). A double-embedding technique using celloidin followed by the typical 

 paraffin method (Johansen) was necessary for some of the tougher structures. 

 Sections were stained with Northen's modification of Foster's tannic acid — ferric 

 chloride method (Johansen) and mounted in Canada balsam. 



Magnification of the figures is indicated by brackets, which represent 100 fx 

 unless otherwise labeled. Where only one bracket appears on a plate, all the 

 figures of that plate are at the same magnification. Various conventions requir- 

 ing explanation have been used in the drawings. In figures showing cellular 



-157- 



