452 



MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



IVol. 9, No. 3 



cate a specialization. The very small and decidedly elliptical grains found by 

 Wodehouse (1929 a) in Trixis and by Erdtman (1952) in Mutisia are likewise seem- 

 ingly advanced. 



The development of areas of thickened exine, seen here particularly in Stiff- 

 tia chrysantha and Duidaea tatei, would seem to be an advanced character. Polar 

 thickenings, such as those seen to a moderate extent in several of the genera con- 

 sidered in this paper, have been reported by Wodehouse (1929 a, b) for a number 

 of genera of Mutisieae which are not regarded as being closely related. Polar 

 thickenings, therefore, may represent a sporadic tendency within the tribe, and 

 may perhaps be regarded as advanced, at least where most conspicuously 

 expressed. 



It is of interest to compare the two-layered ektexine in the Mutisieae with the 

 condition found in the Heliantheae. In Helianthus annuus (illustrated by Erdtman, 

 1952) and other relatively primitive Heliantheae examined by the writer, a thick 

 ektexine consisting of only a single layer but containing small lumina (largely 

 restricted to the spine basis) is found exclusively. It seems apparent that the 

 types of exine stratification found in putatively primitive Heliantheae and Muti- 

 sieae respectively represent two lines. Speculation concerning the relative primi- 

 tiveness of these two patterns would not seem justified because of our extremely 

 limited knowledge of this essentially cytological character. It is of interest, how- 

 ever, that the more complicated ektexine occurs in other tribes of Compositae, as 

 an examination of the drawings of Erdtman (1952) will show. 



II. Floral Venation 



The genera of Guayana Highland Mutisieae treated taxonomically earlier in 

 this series by Maguire et al. (1957) demonstrate a wide variety of floral venation 

 patterns. Among these are the most complicated yet to be reported for the Mu- 

 tisieae, and indeed, the most elaborate yet found in the Compositae at large. Since 

 floral anatomy appears to be of considerable evolutionary significance within 

 the Compositae, a description of these patterns seems desirable. 



Herbarium material was used exclusively for these studies, and the kindness 

 of Dr. Bassett Maguire in providing this material from the Herbarium of the New 

 York Botanic Garden is gratefully acknowledged. Several additional specimens, 

 designated here as (GH), were made available through the courtesy of the Gray 

 Herbarium, Harvard University. For reading the manuscript and offering helpful 

 suggestions, sincere appreciation is expressed to Dr. Reed C. Rollins and Dr. 

 L W. Bailey. 



METHODS 



Wnole mounts of corollas, achenes, and styles were used to a large extent 

 in determining venation. These were prepared by clearing in warm 2.5 per cent 

 aqueous sodium hydroxide until most of the cellular contents had been removed; 

 further clearing was achieved by prolonged treatment in 25 per cent aqueous 

 chloral hydrate. After clearing, specimens were washed, dehydrated in an ethyl 

 alcohol series, stained with a 1 per cent safranin solution in absolute ethyl 

 alcohol, destained, transferred to xylene, and mounted in a xylene-soluble resin 

 ("Dama^'). Because achenes become sclerified upon maturation, obscuring vena- 

 tion, and because the disposition of interior and exterior bandies in the achene 

 cannot be ascertained from clearings, serial sections of flowers were prepared to 

 supplement information derived from the whole mounts. These were made by 

 embedding cleared flowers according to the tertiary butyl alcohol series of 

 Johansen (1940) and staining sections following Northen's modification of Foster's 

 tannic acid-ferric chloride method (Johansen 1940). 



