186 HOW TO STUDY PLANTS. [LESSON 30. 



or banner; 'the two side petals on stems (claws) of their own, 

 and peculiarly shaped, called wings; the two lower petals also 

 clawed and cohering by their edges to form a little boat-shaped 

 body, the keel. Pressing open this keel it is found to contain the 

 stamens and pistil. It will be further observed that ail five petals 

 blend below to form a tube, and thus refer our Clover to Division 

 II., GAMOPETAL^E (p. xii.). This will serve to emphasize the 

 fact that the division of Dicotyledons into Polypetalae, Gamo- 

 petalge, and Apetalse is purely an artificial classification (572), for 

 most of the relatives of our plant belong to the POLYPETALAE, and 

 there our order is placed in the Manual. 



535. Under GamopetaltE we select the second group, B, that 

 with superior ovaries ; and under 13 we choose 1 , in which the 

 stamens are more numerous than the lobes of the corolla, for 

 we find ten stamens and but five petals. Between the two orders 

 left to choose from, the first is evidently the one which accords 

 with our plant, that with " pistil single and simple : leaves 

 compound." 



536. Accordingly we turn to the order LEGUMINOS^E, page 50, 

 and find it subdivided into three suborders. A reading of their 

 characters decides in favor of Suborder I. PAPILIONACE^E. Two 

 choices (marked by asterisks) are now presented, dependent upon 

 stamen characters. Examining our stamens we find that nine of the 

 ten are united by their filaments below into a tube, while the tenth 

 is about free. The stamens, being thus thrown into two groups, are 

 said to be diadelphous (Fig. 227, p. 112). This discovery decides 

 in favor of the group under * # . Under this the reniform an- 

 thers, the three leaflets wh ; ch are toothed, and the flowers in a 

 head (capitate), lead us to the fourth genus, TKIFOLIUM. 



537. Turning to page 54, we find 12 Clovers grouped into three 

 divisions, among which our three leaflets and heads without invo- 

 lucres decide in favor of the second, marked * * . The habit 

 of our plant being plainly caulescent (85), we are left to decide 

 between species 2, 3, and 4. Reading the descriptions of these, 

 we find that our plant does not correspond with any of them, 

 but the reference to foot-note 2 at our group heading leads us to 

 find in this same group two additional species, T. pratense and 

 T. repens, and our plant plainly suits the description of the first. 

 We thus learn that the plant is called Trifolium pratense, and that 

 it does not belong to our native flora. 



