164 



KNOWLEDGE 



[July 2, 1894. 



stems and endorhizal roots ; the leaves of dicotyledons are 

 net-veined, their stems exogenous, and their roots usually 

 develop in exorhizal fashion. The tirst leaves of the 

 dicotyledonous embryo arise in a whorl — they are the two 

 opposite cotyledons ; the subsequently-formed leaves may 

 arise either in pairs or verticils, but far more frequently 

 they follow the I arrangement. There is but one cotyle- 

 donary leaf on the embryo of a monocotyledon ; in 

 succeeding leaves the ^ phyllotaxis prevails, though i is 

 not uncommon. 





e 



I, Monoiotvledonoiis Type. 



V^. C CD S>^ 

 II. Dii'oh ledonous Tvpc. 



The persistence of these characters points to an early 

 separation of the primitive angiosperms into two well- 

 defined groups having their leaves differently arranged. 

 Although monocotyledons appear first in the geological 

 series, it is by no means certain that they represent the 

 earlier type. Pentamerous and trimerous flowers may 

 have arisen independently after the separation of the two 

 families, but the leaf systems they respectively represent 

 should at least admit of reference to a common origin. 

 The opposite and decussate arrangement some botanists, 

 with good reason, regard as primary for dicotyledons ; 

 from it the | and all other divergences occurring in this 

 class can easily be deduced. The two opposite cotyledons 

 favour this view, and the passage from the opposite to the 

 1^ arrangement may be actually observed in the artichoke, 

 willow-herb and other rapidly- growing plants. The 

 ^ type does not admit of direct derivation from opposite 

 leaves, and this Heuslow gives as the reason why it is 

 never found among the foliage of dicotyledons ; the 

 3-merous symmetry of such exceptional flowers as Birbciis 

 he regards as due to the breaking up of a high continuous 

 spiral into groups of threes. The ^ arrangement so 

 characteristic of monocotyledons may have come from the 

 opposite type in this indirect way, through an -^\ or more 

 complex cycle ; but it is much more probable that it arose 

 by symmetrical decrease. A hint of what was, perhaps, 

 the condition in the common ancestral form is furnished 

 by the herb paris, belonging to a family which has the 

 closest relations with the Liliacese. The flowers of Paris 

 ijuailrifnlio are 4-merous ; its leaves are reticulated and 

 arranged in a whorl of four. We have in this undoubted 

 monocotyledon a combination of monocotyledonous and 

 dicotyledonous characters with a phyllotaxis closely approxi- 

 mating to what in all probability was the primitive type in 

 dicotyledons. It should not be forgotten, however, that 

 in the phyllotaxis of fossil cryptogams and gymnosperms 

 much diversity is found. 



In diagrams I. and II. the whorls are shown alternating ; 

 this is their normal position. Now since foliage leaves 

 develop in acropetal succession — that is, from below up- 

 wards — the floral organs ought theoretically to do the 

 same ; but if they followed the strictly spiral succession, 

 the whorls would not be alternate, but superposed. To 

 account for the alternation we must therefore assume that 

 each whorl as a whole has shifted its position, the dis- 

 placement being equivalent to the rotation of a floral axis 

 through half the angular divergence. The angular 

 divergence itself is observed to change in some flowers : 

 in aconite, for example, from f in the calyx to J and 



T/r in the corolla and andrneeium. One or two of the 

 Ranmiculaccte, such as Gandrlla and Helleborus, have the 

 petals superposed to the sepals in strict accordance with 

 spiral phyllotaxis, but this is rare in the floral envelopes. 

 Where the andrn'cium consists of numerous stamens, 

 these are frequently arranged in a spiral manner, giving 

 rise to superposed whorls. The sepals and petals of the 

 buttercup arise in alternate whorls, but the stamens and 

 carpels develop in spiral fashion like ordinary leaves. To 

 this condition the name hemicyclic has been given. In the 

 water-lily order, Nymphaceff, all parts of the flower follow 

 the spiral order ; this condition also occurs in the camellia, 

 in the Magnoliacea' and Calycanthacefe, and is approached 

 by several of the Ranuuculaceff. A significant fact in 

 connection with the simplicity of the flower of the water- 

 lily is the circumstance that, notwithstanding its truly 

 dicotyledonous embryo, the root-stock shows the endogenous 

 structure of the monocotyledonous stem. 



Considerations like the foregoing make it clear that 

 acyclic flowers, or those whose parts form a continuous 

 spiral, represent a primitive type, upon which the hemi- 

 cyclic condition of the buttercixp is a slight advance. The 

 passage from these to the regular alternating whorls of the 

 eucyclic class involves a modification of the phyllotaxis for 

 each succeeding whorl. Under the term cyclic are included 

 all flowers whose parts occur in whorls, but some confusion 

 has arisen from a lax use of the latter term. Some 

 writers call any set of leaf organs a whorl which arise on 

 the same horizontal zone of the axis — i.e., which are pro- 

 duced at the same height, or what amounts to the same 

 thing, at equal distances from the growing point ; others 

 restrict the term to circles in which the parts appear 

 simultaneously. Sepals, as a rule, arise in succession ; 

 petals, for the most part, simultaneously. The calyx of 

 the rose illustrates this successive character ; its outer- 

 most and oldest sepal has fringes on both its edges which 

 are free ; so has the second sepal, which is placed at an 

 angle of 144° to the first — i.e., with a divergence of |, . At 

 an equal distance from the second stands No. 3, fringed on 

 its outer edge only ; No. 4 is similar, while No. 5, which 

 completes the cycle, has no fringes, both its edges being 

 overlapped by the other sepals. A corresponding order 

 can often be traced even in the corolla — in the butterfly 

 blossoms of Leguminosae, for example, the large standard 

 petal begins the leaf cycle, one of the keel petals com- 

 pleting it ; and generally the ;estivation of the flower — 

 that IS, the manner m which its parts are disposed in the 

 bud before expansion — admits of explanation on the prin- 

 ciples of phyllotaxis. The two examples just given 

 illustrate respectively the quincuntial and vexillary modes 

 of iestivation. As early stages in the evolution of the 

 blossom, then, we have to note ; I. The spiral arrange- 

 ment of the floral organs. II. Whorls due to the arrest 

 of the internodes of the floral axis. III. Alternation, or 

 a change in the orientation of each whorl disturbing 

 the spiral order ; and IV. Simultaneous whorls with 

 synchronous development of parts, which still further 

 obscures the original phyllotaxis. 



NATURE'S PROTEST AGAINST CHANGE. 



By Vaughan Coknish, M.Sc, F.C.S. 



THE two great laws of conservation — the law of con- 

 servation of mass and the law of conservation of 

 energy — have familiarized the educated world with 

 the doctrine that in Nature nothing is lost. The 

 form of matter changes, but the quantity remains 

 unaltered ; the form of energy changes, but its ' ' mechanical 



