1 he Malta or Portugal Oranges. 21 



adhering but lightly to the pulp, I have seen what I 

 considered an undoubted citron tree, with white 

 flowers ; the sweet lemon I have never seen but with 

 white flowers ; the khatta, which has most characters 

 of an orange, is usually with a pronounced nipple ; 

 almost all the keonla oranges have a flat nipple ; the 

 keonla of Aurungabad, pi. 125, fig. a, the mussembi 

 of Poona, pi. 44, fig. f, and many specimens of the 

 Seville orange, are decidedly warty ; and many 

 varieties of the Malta orange in the English market 

 have their skin so adherent to the pulp that it is next 

 to impossible to remove it without cutting it with a 

 knife. 



The fact is, it is rather puzzling sometimes to have to 

 decide which is an orange, which a pummelo, and which 

 a lemon, or citron. I am afraid it has been taken too 

 much for granted that the lemon must be yellow, 

 elongated, and with a mammilla and red flowers, and 

 that the orange must be red, spherical, without a 

 mammilla, and with white flowers. There is no good 

 natural reason for this. It would appear much like 

 Linnaeus's classification of plants by the number of 

 stamens. 



The question whether the bitter and the sweet 

 orange are merely varieties, one of the other, or 

 distinct species, has not been satisfactorily answered. 

 I would here ask which sweet orange is meant ? The 

 Malta or Portugal, and its varieties ; or the suntara or 

 kamala and their varieties ? I have elsewhere shown 

 that there are two distinct types of sweet oranges, 

 which may owe their parentage possibly to distinct 

 wild kinds. The attempted replies, however, to De 

 Candolle's question are very unsatisfactory, and he 

 considered them contradictory. He quoted from 

 Gallesio's " Traite du Citrus " : " I have during a 



