NOTES ON SAPOTACE.E. 357 



Of the remaining Mimusopem, including the rich Mascarene 

 collections in the Paris Herharium, I have analysed nearly all the 

 species, with the following results : — 



Imhricaria coriacea, DC., is identified by Bentham and Hooker 

 with the tyi^e and sole s^Decies oiLahramia, DC. ; I find, on dissecting 

 flowers of a large number of specimens, that the staminodes, instead 

 of being absent as hitherto described, are well developed, deltoid- 

 subulate, shaggy behind. Hence this plant is only separated from 

 Imhricaria by the 3-merous flower and the pleiomerous gynasceum. 

 But 3-merous and 4-merous flowers exist side-by-side in Mimusops 

 (on the same branch in M. lioxhurghiana, Wt.). And pleiomery of 

 the gynaeceum, not held of sufficient importance to separate 

 Chrijsoplujllum. Caiviito from the 5-carpellary sj)ecies of Chryso- 

 jjhijUiim, cannot be of more weight here. I have, too, found an 

 undoubted Mimusops of Boivin's from Zanzibar, which is a true 

 Mbiiusops iu every other respect, but possesses a 12-merous 

 gynaeceum [M. annectens, mihi, in Herb. Mus. Par.) Again, 

 Lahramia is confined to Madagascar, and, with one exce^Dtion, 

 Imhricaria is limited to Madagascar and the Mascarene Islands. 

 Therefore Lahramia must lose its autonomy. 



Now, I turn to Imhricaria itself: I found, in Paris, a beautiful 

 specimen of Chapelier's from Madagascar {Mimusops Chapelerii., 

 mihi), which has the flowers of a " Lahramia,'" save for the 

 petaline appendages, which are quite entire, as. in Mimusops. 

 Where is this to be placed under the existing nomenclature ? 

 I have just stated that Imhricaria is, with one exception, 

 Mascarene, using the word iu a wide sense. On turning over the 

 species in the Kew herbarium, I find that Mimusops Cruegeri, 

 Grisebach, from the Antilles, has the lobed x^etaline aj)pendages of 

 Imhricaria, and, indeed, has the 4-merous flower and habit of one. 

 Can the entire or lobed form of such secondary structiu'es be alone 

 used to form a genus ? I think not. 



Si/narrhena was founded by Fischer and Meyer for two South 

 American species, of which the only distinctive character is the 

 connation of stamens and staminodes at their base into a tube, 

 distinct for a short distance from the corolline tube, to which, 

 however, it is adnate. It has been reduced to a section of 

 Mimusops by all subsequent authors. 



Finding that two 3-merous species of this group, distinct from 

 J/i??n(soy^s proper only in the fertility of the alternipetalous stamens, 

 had been confounded with Lahourdonnaisia, I regret to say that I 

 formed them into a distinct genus under the name Eichleria, 

 afterwards altered to Muriea. But of the two species, one is 

 from Natal, the other from the Antilles. Besides, they are of very 

 difi'erent habit. I now therefore regard them as independent 

 reversions to a diplostemonous type, such as I have found in a 

 flower (normal in other respects) of a specimen of Sideroxylon 

 sandu-icense, A. Gray.'-' What was monstrous has become normal. 



* Of staminodes. one was petaloid normal, another irregular in shape and 

 bearing an anther ; the other three converted into true stamens like the anti- 

 petalous. 



