SAFFORD-—CLASSIFICATION OF ANNONA, 15 
considerable variation in the size of the leaves and the flowers, but a careful 
comparison of them shows that the “ alligator apple” of the Bahamas (Annona 
glabra L.) and the “water apple” 
of Jamaica (Annona palustris L.) 
are indistinguishable, a fact which 
was recognized by St. Hilaire, 
who, in company with Desfontaines, 
made a careful comparison of the 
two. In referring to the confusion 
of the two species St. Hilaire says: 
“Anona glabra, then, is still a spe- 
cies as little known as in the time 
of Linneus. That immortal nat- 
uralist cites no other synonym of 
his species than the phrase and f[F1e.19.—Flower of Annona glabra, showing 
figure of Catesby, in whose work wnat nner petals. One petal removed. 
only a vague notion of the plant in ° , 
question can be obtained. Lamarck, while copying Catesby, added to that 
author’s description features borrowed from other species. As for ildenow, 
he did nothing more than copy Linnus, and contented himself with adding to 
the synonym from Catesby another synonym 
borrowed from Duroi. Moreover, it is quite 
evident that he had not time to read the text 
of the Harbkesche Baumzucht; for its author 
states that he based his description of Anona 
glabra on a young plant sent from England 
under that name, but of which he saw neither 
the flowers nor the fruits; and as he adds that 
this young tree growing in a pot had dentate 
leaves, it is certain that the plant was not even 
an Anona. It is perfectly evident that if Wilde- 
now had taken the trouble to read all the de- 
scriptions cited by him, he would never have 
reached the end of his first volume, and the 
impossibility which one finds of verifying so 
many obscure synonyms shows how necessary 
it is for botanists to free themselves from the 
obligation of citing them. To indicate them 
without investigation is but to multiply and 
perpetuate errors; to verify them is fruitlessly 
to spend time which would be more usefully 
Fia, 20.—-Gynecium of Annona €mployed in observing.” * 
glabra, showing consolidated Included in this section and closely allied to 
ovaries, Styles borne on the  Anona glabra, if not identical with it, is A. 
ovaries ; two. stamens at the  }.1ainii Pierre, of the west coast of Africa, ab- 
cs ° normal 8-petaled flowers of which misled Pierre 
as to its relationship to the typical A, glabra, which is found on the adjacent 
coast. 
Section 8. ATRACTANTHUS sect. nov. 
Type species, Annona acutiflora Mart. (pl. 22). 
In this section the flower bud is acuminate, The outer petals are united at 
the base but open widely when the flower is mature (fig. 21). The inner petals 
* St. Hilaire, Augustin de. Plantes Usuelles des Brasiliens 30: 5, 1824, 
