clxx PROCEEDINGS OF THE ROYAL HORTICULTURAL SOCIETY. 



of this plant in process of germination, with the following remarks : " It 

 will be seen, as I stated in the communication I sent in June, that more 

 than one cotyledon issues from some, if not from all, of the ' seeds ' or 

 fruits, and on dissection more than one embryo or ovule will be seen. 



" The nine ' seeds ' I sowed last year produced respectively 1, 2, 2, 3, 3, 

 4, 4, 4, and 6 bulbs, some of which came up much later than others. It 

 is possible that some of these might be offsets, something like Tulip 

 1 droppers,' but on taking them up there was no sign of there having been 

 any connection between them. The three one-year-old bulbs enclosed 

 came from one 'seed' (which had four bulbs altogether). 



"I think it may also be noted that though the cotyledons begin to 

 grow within the fruit in various directions, as you will see on dissection, 

 they all eventually issue from the underside of the fruit as it lies on the 

 soil, irrespective of the position of the fruit itself, for they were purposely 

 placed with different sides downwards. 



" The ovules within the fruit do not germinate simultaneously, and 

 there is apparently competition between them for the substance of the 

 fruit ; the ones that germinate first apparently produce larger bulbs than 

 the later ones. 



" I think it will also be seen that the tip of the cotyledon is a true root. 

 This is apparent even when the cotyledon is only J in. outside the fruit ; 

 when it is from about J to j in. long it develops a little tuft of bristles or 

 root-hairs at the junction of the neck of the cotyledon and the root, by 

 means of which it anchors itself in the soil, and then the root grows on 

 downwards. This takes place when the soil is kept moderately moist, 

 but I suspect that if the soil were drier the cotyledon would extend itself 

 more before anchoring and sending down the root. 



" That the process issuing from the fruit is a true cotyledon is more 

 apparent at a later stage, when the lower part has thickened into a bulb 

 and the tissue of the fruit all been absorbed by the young seedling. Then 

 if the semi- withered, semi-rotted upper part of it (coiled or doubled up 

 inside the remains of the fruit) is dissected away, the dormant plumule 

 will be seen issuing from the centre of the fresh, somewhat thickened 

 part left. 



" I have no doubt you will observe this and more with the seeds I send 

 you, but I will also send some more for the committee to see when in 

 this later stage. 



" I have a Himantophyllum seed which has produced two seedlings 

 (now in their first leaf and both still connected to the seed). This seems 

 to me to be more possibly a case of two embryos in one seed, but my 

 ' further observations this year have strengthened my conviction that with 

 HymenocalUs Harrisiana it is a case of several seeds or ovules in a 

 modified pod — or fruit." 



I h\ Rendle reported on this communication : " The 1 seeds ' of Hi/meno- 

 callis Harrisiana, from which several seedlings are growing, and which 

 Mr. Bliss suggests are 1 several seeds or ovules in a modified fruit,' are an 

 example of the bulbiform type of seed frequent in AmaryllidacecB. They 

 are polyembryonic, but in the advanced stage of development which they 

 have reached it is impossible to suggest how the polyembryony has arisen 

 in the seed. 



