lO 



BULLETIN OF THE LABORATORIES OF 



Fig- 3- 



simple and numerous, when the pistil is compound there is pre- 

 sented a circle of carpels showing more or less perfect coales- 

 ence. It is by a possible combination of these two principles 

 that there is produced the peculiar form found in the navel 

 orange, figure 3, where the plurilocular 

 pericarp contains a second circle of lo- 

 culi bearing a separate style. It is evi- 

 dent that this second included fruit 

 must represent a continued develop- 

 ment of the floral axis but whether that 

 continued development should repre- 

 sent a pistil and so a circle of carpels 

 may be questioned. The fact that in 

 this particular variety of oranges the seeds are rare and often 

 entirely wanting, as they were in the specimen studied, would 

 seem to indicate rather that these accessory parts are closely re- 

 lated to the aborted seeds. It would appear then that in the 

 orange and so probably in the citrus family the seeds are to be 

 considered as lateral developments from the central floral axis 

 and the carpels entering into the compound pistil as sterile. 

 The thought suggests its self that the development of the pulp 

 from the innermost layer of the pericarp in the form of isolated 

 lobes of tissue, into the plurilocular fruit, may take the place 

 of the more usual form of seed development from the margin or 

 surface of the carpel leaves. 



* * :^ 



* 



The morphological relations of the ovules is pretty clearly 

 shown by the following interesting monstrosity in a single tulip: 



Figure 4 is a full sized sketch of the pistil, showing two 

 faces of its triangular form. 



Figure 5 is the top view of the stigma, which is sessile. 

 The two portions which are contiguous belong to the same 

 carpel as the ventral suture is in the middle of the face and 

 the dorsal at the angle of the ovary. 



