STRUCTURE.] IN POPPYWORTS. 377 



We ought not indeed to be surprised at coming to this 

 result; for if the ovules are, as botanists generally believe 

 them to be, a modification of buds, then the uncertainty in 

 the position of the placentary lines will only be conformable 

 to the uncertainty in the origin of buds from leaves. If in 



opposition, forming a furrowed stigmatic line, which necessarily corresponds to 

 the placentas, and consequently alternates with the valves. 



In Chelidoniura the stigmatic lobes (which in Macleaya were capable of separa- 

 tion and complete reflexion) are more rigidly erect, but the furrowed line bears 

 the same relation to the placentas, which now exhibit their bi-carpellary origin 

 by bearing a double row of ovules. 



In Glaucium. the stigmatic lobes become enlarged, but otherwise remain as in 

 the last example. The parietal placentas are furnished with a linear spongy 

 growth projecting from between the rows of ovules of each placenta, and uniting 

 with that of the opposite side in the centre of the capsule, which is thus con- 

 verted into two cells. This spongy dissepiment is usually described as arising 

 from the extension of the placentas ; an attentive examination at different 

 periods of growth, however, will show that it is really distinct in structure 

 though attached to them. 



In Hunnemannia we have the first indication of an addition of parts ; the 

 stigma being obscurely four-lobed, indicating the manner in which new carpels, 

 will, in other genera, become interposed between the two primary ones, which 

 alone exist in the preceding instances. 



In Eschscholtzia the additional stigmas (which are only indicated in Hunne- 

 mannia) are considerably developed, but are separate from the primary ones. 

 The fact of their being the superadded stigmas is, however, indicated by their 

 being shorter than the others. 



In Meconopsis the additional carpels (only sketched forth and indicated, as 

 it were, by the additional stigmas in Hunnemannia and Eschscholtzia) are per- 

 fected, each carpellary valve contributing by its margins to the formation of two 

 parietal h'near placentas, which latter correspond with the stigmatic rays. Each 

 stigmatic ray is formed precisely similar to the stigma of Macleaya, Chelidonium, 

 and Glaucium, being furnished with a central depressed line, indicating its forma- 

 tion from the union of the corresponding halves of the two contiguous carpels. 



In Argemone the radiated stigma presents an undulatory folded appearance 

 in consequence of the increased growth of the intervening tissue, which in the 

 preceding genera (excepting Eschscholtzia) separates the lateral portions of the 

 stigmatic extremity of each carpel. 



We now arrive at Papaver, in the different species of which the capsule 

 presents several states of complication by the successive addition of a greater 

 number of carpels, which in P. somniferum sometimes amount to sixteen. The 

 parietal placentas, which in all the preceding genera are linear, now project in 

 towards the centre of the capsule, partially dividing it into as many imperfect 

 cells. The stigmatic rays, which, as in the preceding instances, are equal in 

 number to the placentas, and opposite to them, are, as already described, double, 

 and only differ from those of Argemone in having the intervening tissue, which 

 separates the two margins of the stigmatic extremity of each carpel, plane 

 instead of folded." (Annals of Natwal History, Vol. x.). 



