THE FAMILY CHAIN 355 



dehisce b}' longitudinal slits. The expression T ^ means that the 

 torus is con^'ex and implies that the perianth and andro'cium are 

 hypogj'iious. When as in Psonia they are perigynous this is indi- 

 cated by T ^' which represents the torus as concave. 



The form of the ovules is shown by a mark placed over their 

 numerical sign, a circumflex accent-mark meaning that the ovule 

 is anatropous. Their ventral position is understood in simple 

 pistils, while in compound pistils like that of Nigella, the single 

 parenthesis after the number of carpels implies that the ovules are 

 on an axile placenta. 



When the pericarp becomes fleslij' as in Actaa this is indicated 

 by an exclamation mark after the C. When the pericarp is drj^, 

 a.s in Caltha, there is instead an inverted exclamation mark. Inde- 

 hiscence is indicated by the sign < . When the pericarp dehisces 

 along a ventral suture as in Caltha, etc., the sign < is employed. In 

 all the formulas the expres.sion G-N implies that the embryo is 

 uncoiled within albumen. 



The scheme of plant formulas which is here proposed and which 

 will be further elaborated in the following pages, is an extension 

 and modification of the floral formulas used by many botanists. 

 As a sort of botanical shorthand of wide application it is bclie^'cd 

 that the student will find it not only lalaor-saving but helpful in 

 grasping plant relationships. After a little use, what seemed strange 

 will have become familiar and a glance will discover important 

 characters that might easily escape notice in comparing equally 

 full verbal descriptions. 



105. The family chain. Having learned the signification of 

 these symbols we are now in position to use the formulas as a readj' 

 means of comparing the main structural features of our representa- 

 tive genera to see how they are hnked together. Take, for instance, 

 Caltha and Pseonia. If we conceive of a marsh-marigold having a 

 concave torus, a perianth differentiated into calyx and corolla, and 

 pinnately compound leaves, such a plant would be classed as a 

 peony. B}^ these same features, however, it might be distinguished 

 from all the other genera. Therefore, although closel)' linked with 

 Caltha, Peeonia is placed on a line apart in the tabular view. 



Helleborus differs from Caltha chiefly in having the carpels some- 

 times coalesced and in possessing staminodes. In these respects 

 it is a link connecting Caltha with Xigella which has the carpels 

 always coalescent, and differs from Helleborus only in having 

 pinnate instead of palmate lea^-es, some of which may be so near the 

 flower as to constitute an involucre, and in consisting of annual 

 rather than perennial herbs. 



Aquilegia, with its carpels distinct, is more like Caltha, but 

 difTers from both Caltha and Nigella in having the carpels always 

 five, staminodes in two inner sets of five and one outer set of the 

 same number, and in having the leaves ternately decompound. 



