172 



HARD WICKE ' S S CIENCE- G SSIP. 



Pistils : numerous, forty or more, forming a hollow 

 cylindrical-shaped body, enclosed by the column and 

 surrounding the carpels, which are arranged in a 

 circle as shown in Fig. 86. Tn the single hollyhock 

 the pistils, as to their upper portions, are generally 

 dark crimson, the petals assuming the same colour, 

 whilst the stamens are bright yellow ; the column, 

 however, is variable, sometimes being crimson, some- 

 times yellow. In the double hollyhock the column 

 and base are usually yellow, the stamens the same : 

 the pistils, however, vary, sometimes being yellow, 

 but as frequently the same colour as the petals, 

 though generally of a lighter shade. 



Stamens : spring in considerable numbers from 

 the top and sides of the column. In the double 

 flower they are interspersed with the centre florets. 



In observing a large number of single as well as 

 double flowers, I noticed that the pistils do not 

 appear above the column for some days after the 

 flower opens and the anthers have matured, thus 

 favouring cross-fertilisation. When fully raised they 

 bend outwards — more in the single than the double, 

 because in the latter they are obstructed by florets — 

 and become intimately mingled with the upper 

 stamens, or the stamens and florets, as the case may 

 be ; but I doubt if this is often the cause of self- 

 fertilisation, as it seems to me that fertilisation 

 usually takes place directly the pistils appear, and 

 before they have had time to bend ; that is, if the 

 proper agents have satisfactorily done their work. 

 The habit may, however, be useful as a sort of 

 guarantee against possible failure, by bringing the 

 pistils and anthers so close together that fertilisation 

 might take place the more readily by means of the 

 visiting insects, or even without their co-operation. 

 In the figures the pistils appear separated for only 

 part of their length ; but, as a fact, I believe they 

 are distinct from each other throughout in most 

 plants, and their number, as far as I could ascertain, 

 invariably corresponds with the number of the 

 carpels. 



The centre florets of the double flower (see Fig. 85) 

 exhibit, in a wonderful manner, the various steps 

 between stamen and floret. The figure represents 

 a vertical section of the flower, with all but a few 

 florets removed. One floret is cleft at the top, but 

 otherwise, in shape like an ordinary petal ; another 

 is a half floret, with a projection on one side. The 

 next shows a perfect anther in place of the pro- 

 jection ; another is divided, part being floret, part 

 filament and anther ; another looks like a shrivelled 

 floret, with the top turned over. 



Of course it may be urged that any particular 

 intermediate form never itself becomes the service- 

 able shape to which it is apparently tending, as, for 

 instance, a petal, a floret, a stamen, or a pistil, and 

 this is undoubtedly true ; but it is also equally true that 

 beneficial modifications are reproduced in each suc- 

 cessive plant until they do reach the goal at which 



they aim. So if we could follow through several 

 points the floret with a simple projection, we might 

 trace it upwards to a stamen or downwards to a 

 floret, the projection gradually assuming the form 

 of a stamen or the second half of a floret, as the case 

 may be, according to the direction in which the 

 modification had set in. 



Assuming that the double hollyhock is derived from 

 the single, and I suppose there cannot be two opinions 

 about this, it follows that the florets trace back to an 

 ordinary stamen. All the evidence I have adduced 

 points in that direction, and it is no more unreasonable 

 to believe than that a stamen should be evolved from a 



Fig. 85. — Vertical Section of Double Hollyhock, showing 

 a few of the central florets. 



Fig. 86. — Pistils, carpels, and inner bracts, 

 the last turned down. 



petal. The modification of parts is incessantly going 

 on, and a particular organ is not always reached 

 by direct steps ; thus I can understand a petal first 

 becoming a stamen, and then a floret by evolutionary 

 process. It may be called reversion ; but whatever 

 term is most applicable to the phenomenon, the case 

 now under consideration shows how marvellously 

 the various parts of a flower can mould themselves 

 to their requirements, applying a gain, or fitting a 

 loss to some useful purpose, and how they can 

 expend lavishly or retrench carefully without being 

 charged with unnecessary extravagance on the one 

 hand, or abortive penuriousness on the other. With 

 plenty and to spare they will support a great number 



