80 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
seen, on the other hand, that the tree was frequently described from the West Indies 
in the course of the eighteenth century. It may have reached Cuba from Yucatan 
in prehistoric times and spread from there to the other islands. 
Although it is seldom met with in Central America and Mexico above 1,000 meters, 
the upper limit of the tierra caliente, the sapodilla tree reaches far up into the tem- 
-perate belt of Colombia and is even grown around Quito, in Ecuador, at an altitude 
of about 2,800 meters. . 
The adult trees seem to vary greatly in size according to locality. Cook and Collins! 
give 7 to 9 feet (2 to 3 meters) for its stature in Porto Rico; Jacquin * gives 10 to 15 
feet (3 to 5 meters); P. Browne ® says that it “‘rises to a considerable height.”” In 
Guatemala and Colombia I have often seen specimens 18 theters high and over. As 
a general rule, it seems that the tree is of a lower stature in the West Indies. The 
specimens seen by me in Port Limon and on the plains of Santa Clara in Costa Rica 
were also of less size and more densely foliated than those on the Pacific coast. This 
can, however, be accounted for by differences of age, climate, and other local con- 
ditions. 
Pierre ¢ has described several varieties which have not been found among the 
numerous specimens investigated in connection with the preparation of this paper.® 
Frequent and considerable variation was noted in the relative length of the calyx 
and corolla, or of the latter’s lobules and the staminodes, these being in most cases 
adherent to the lobules and not free above the insertion of the stamens as represented 
by Engler.® With reference to the general form of the flowers an old observation of 
Loefling was confirmed and thus quoted by Jacquin: 7 ‘‘Flores inodori, corolla albida, 
diu persistentes. Hi ante fecundationem figuram habent ovatam, in ipso autem 
fecundationis actutoti explicantur magis, ut evadant campanulati; quod, ut ista 
succedat, antherae inclusae stylusque corolla longior videntur exigere: unde tunc 
in situ figuraque mutatio partibus accidit insignis. Fecundatione auctem peracta, 
ovatam denuo assumunt. Extra hunc actum florem descripsisse videtur beatus 
Loeflingius, cujusmodi ipse illum ego saepissime examinavi: addidi igitur charac- 
terem, qualem in ipsamet fecundatione semper se mihi exhibuit.”’ , 
The seeds also vary widely, not only in number but also in shape, as can be seen 
from the accompanying plate. As to their number, I found that in the Cauca Valley 
it is usually not over three and very often less, and it was with no little surprise that 
I saw later in Velasco a reference to the Ecuador fruit as being also 3-seeded, while 
my own experience in Central America, as well as that of most authors, indicates a 
larger number. On the occasion of a recent trip to Venezuela, where the fruit is a 
1 Contr. U. 8. Nat. Herb. 8: 66. 1903. 
2 Loc. cit. 
3 Civ. Nat. Hist. Jam. 200. 1789. 
4In Urban, Symb. Antill. 5: 97. 1904. 
5In the course of my recent investigation of the flora of Panama, I discovered a 
remarkable variety of this species at Patifio, on the southeastern shore of San Miguel 
Bay. The tree is rather small, not reaching over 8 meters high; the trunk is 
straight and the branching divaricate, almost horizontal, with the lower limbs 
drooping; the crown, the lower part of which is only about 1.5 meters above the 
ground, is regular and oblong-elongate. The tree was loaded with fruits, these forming 
dense clusters at the end of the branchlets. The peduncles are 1.5 to 2 meters long; the 
berries themselves not over 4 cm. in diameter and 3.5 cm. long, globose-depressed 
in shape; the scaly skin is gray, the mesocarp greenish yellow, and the seeds, usually 
5 to 7 in number, always without rostrum. According to the information obtained at 
the place, the tree is commonly found in the surrounding woods, and goes under the 
name of ‘‘nispero de monte.”’ 
6 Pflanzenfam. 4!: 197. 1889. 
7 Stirp. Amer. 58. 1763. 
