2 TEXAS RESEARCH FOUNDATION: CONTRIBUTIONS, VOLUME 5 
tions in British Honduras, Guatemala and Mexico, which have been helpful in 
interpreting certain species. The genus is best represented in the mountains of 
Central America, but with species ranging from Mexico to Peru and into the 
West Indies. Eighteen of the species are known only from the type collection. 
My concept of the species is essentially the same as that of Mez. The accumu- 
lated collections now available are so much greater that a clearer interpretation 
of those known to him is possible. Although some problems have been clarified, 
many new ones have arisen which only future workers can elucidate as material 
accumulates and studies progress. 
The characters used in separating the species of Parathesis have been tested, 
and they are reliable to a high degree. Some species stand out, others are plastic 
and difficult to tie down. In one species the characters may be stable, in another 
exasperatingly variable. Hence the use of such words as “usually,” ‘‘eommonly,”’ 
“generally,” and so forth are essential. They need not impair the worth of the 
key or description, provided that text, illustrations, keys and annotated speci- 
mens are used in conjunction when identifications are made. 
No attempt should be made to use this monograph without first taking the 
time to become acquainted with the fine variations in morphological characteris- 
tics of Parathesis. Expecially is this true of the types of pubescence. The use of a 
microscope is imperative, for all descriptions and interpretations were made with 
floral and other parts magnified from ten to twenty times. 
Nores ON THE GENERAL MorpnHouoGy or PARATHESIS 
The Myrsinaceae of the New World are woody plants. In Parathesis the 
species range from low shrubs to trees up to 20 meters high. The stems are 
indeterminate or determinate. The leaves are alternate, petiolate and pinnate- 
veined. The decurrent base of the leaf blades gives some the appearance of being 
subsessile. Margins of the leaves are either entire, crenulate, serrulate, dentic- 
ulate or coarsely and irregularly dentate. In this treatment of Parathesis, very 
little weight is given to leaf venation. As in all Myrsinaceae, punctate glands 
are present in the leaf blades and these are of varying form and abundance, 
but they are used very little for diagnostic purposes in this monograph. 
The inflorescences are either terminal or axillary panicles, with notable ex- 
ceptions where both types occur occasionally in the same species, sometimes 
on the same limb. In those that are axillary, the terminal bud is conspicuous, 
usually elongate. In P. acuminata Lundell, only two specimens have been seen, 
one of these has the typical terminal panicle, the other a single axillary in- 
florescence. In P. cubana (A.DC.) Mol. & Gomez Maza, the terminal is usual, 
but leafy axillary panicles occur occasionally, and the type of P. corymbosa 
Hemsl., a synonym, is in this category. In P. glabra Donn.Sm. all specimens 
examined have axillary inflorescences except one, and this collection (White 
156, MO), has both types on the same branch. In P. tomentosa Lundell, the 
inflorescences are branched to base, with some axillary, others terminal. This is 
likewise true of P. subcoriacea Lundell. In spite of the exceptions, the division 
of the genus into two sections, species with terminal or axillary inflorescences, 
is the most practical. The terminal are invariably pyramidal panicles. 
The panicles are pinnately branched and variable in size, mostly long pedun- 
culate and usually with slender elongate branches. In some species the branches 
are reduced. In several species the peduncles of the corymbs and umbels are 
