ON CHORISIS. 385 



of these two circles are also opposed to each other. The case of the 

 group of stamens with the petaloid scales behind it in the American 

 Linden seems very closely to resemble that of the clusters of stamens^ 

 in that instance coherent, of Malvaceae in a half-double flower ; of the 

 latter we have the separate petals partially developed as clusters of 

 stamens, and we observe that they are not flat or merely curved, but 

 nearly funnel shaped or folded round again. Let a small portion as- 

 sume the leaf-like aspect and the rest subdivide into separated stamens 

 and we have a remarkable instance of collateral chorisis in an organ so 

 curved in figure as to produce the very appearance exhibited. 



These examples probably include all the varieties that would afford 

 anything special from which to reason, and further details would be 

 unsuitable in this place. I conclude, 1st, that chorisis or the division of 

 a single organ into two or more similar, or approximately similar ones, 

 is a possible and reasonable supposition, and accounts well for a class 

 of facts which the laws of structure previously established did not pro- 

 perly reach. 2nd, that chorisis does not admit of being divided into 

 two kinds, collateral and transverse ; that the latter kind as explained 

 by Dunal, to whom we own the theory, is liable to most serious ob- 

 jections, and is not justified by any facts necessarily implying it, or 

 strictly analogous with it ; that the explanation adopted by Dr. Gray 

 takes the case entirely out of the formation of separate organs from a 

 single one ; and that oppositeness of parts in adjoining circles is no in- 

 dication of those parts being of common origin or belonging to a single 

 organ, so that transverse chorisis may be entirely set aside. 

 3dly. That the ingenious and distinguished authors who have pro- 

 posed and defended the law of chorisis have been led to apply it in 

 various cases which do not really come under the law, and are better 

 explained on other principles, particularly that there is no chorisis in 

 Brassicaceous flowers, and that a number of organs really derived from 

 several distinct circles may be so pressed together as to form one 

 " apparent circle, the parts even being connected by a common expansion 

 derived from the torus, so that a number of crowded parts however 

 regularly set is no proof of chorisis. 



With these restrictions I receive chorisis as an additional principle 

 in the structure of flowers, afi'ording us valuable assistance in bringing 

 them all, however varied, within general rules, and manifesting their 

 common relations. 



