1960] 



MORPHOLOGY AND RELATIONSHIPS OF DIOMMA 



41 



from Manaus to the lower part of the Trombetas River north of the Amazon 

 River, and south of the Amazon River from Juruti Velho and Manes to the 

 northern part of the Mato Grosso and the Territory of Guapore. 



Kranse (1914) placed Sohnreyia in the Rutaceae, in an isolated position, 

 mentioning- that it could be related to Dictyoloma, but lacking mature fruits no 

 definite association could be made. Engler (1931) agreed with Krause regarding 

 the definitely rutaceous nature of Sohnreyia and erected a special monotypic 

 subtribe Sohnreyiinae under the tribe Todclalieae of the subfamily Toddalioidecu 

 to contain this genus. 



METHODS AND MATERIALS 



The investigation undertaken herein depended upon an examination of sev- 

 eral kinds of materials, some by more than one method. Most material was in 

 dry form, but some formalin-acetic-acid-alcohol preserved material was also 

 available for study. Nodal regions and vasculation of petioles in Diomma were 

 examined using a supersaturated solution of phloroghicin in water to demon- 

 strate the presence of lignin. Freehand serial sections were cut directly into the 

 phloroghicin solution, mounted on a slide, and 2 or 3 drops of concentrated HC1 

 applied directly to the section. The appearance of a red-violet color indicated 

 the presence of lignin (vascular tissue). Secretory structures, in leaves of 

 various specimens examined, were demonstrated by the use of small portions of 

 leaflets cleared by treatment with 3 per cent NaOH at 40°C. Subsequently to 

 washing with water, cleared leaves were stained in safranin, dehydrated, treated 

 with toluene, and mounted on slides in Canada balsam. This same method was 

 cm ployed with pistils and other floral parts. Leaves were also embedded in 

 paraffin, sectioned transversely and paradermally, stained in Heidenhain's iron- 

 alum haematoxylin and safranin, and mounted on slides in balsam. Wood speci- 

 mens were prepared for microscopic examination in the conventional way. Small 

 blocks were boiled in water to exclude air and soften, after which they were 

 dehydrated and embedded in celloidin. After being cut on a sliding microtome, 

 sections were stained in Heidenhain's iron-alum haematoxylin and safranin and 

 mounted on slides in balsam. Maceration of wood was accomplished by Jeffrey's 

 method as outlined by Johansen (1940). Macerated wood was stained in safran- 

 in, dehydrated with tertiary-butyl alcohol, and infiltrated with Canada balsam 

 before being mounted on slides. Studies of floral morphology were based largely 

 on boiled flowers from herbarium specimens. Some preserved flowers were also 

 at hand. 



Terminology used in the section on xylem anatomy follows that recommended 

 by the Committee on Nomenclature of the International Association of Wood 

 Anatomists (1957). Terms referring to cell size are taken from Chattaway 

 (1932). Measurements of vessel element and fiber-tracheid length and tangential 

 pore diameter are based on 50 samples each. Only the means and ranges for 

 these measurements are presented because the small number of specimens at our 

 disposal would make meaningless any more detailed treatment of the data 

 (Stern & Greene 1958). 



Specimens examined: Diomma ulei Engler ex Harms: E. Vie 8646 (Type, G; Isotype, L, 

 K); J. A. Steyermark 58982 (F, NY), 59798 (F). D. fruticosa Steyermark: J. A. Steyer- 

 mark 60820 (Type F sterile), 60327 (" Co-type" F fruits); T. Lasser 1786 (US); B. 

 Maguire $ J. J. Wurdack 33808 (NY). Diomma species A: J. A. Steyermark 75699 (NY): 



