112 



♦ KNOWLEDGE ♦ 



[FKa 23, 1883. 



circle will not be \-isible. This arises from the fact tliat a 

 portion of the retina is not sensitive to light. The invisible 

 portion is so large that it will suflice to prevent a man's 

 face from being seen at six or seven feet distance. 



When looking for very minute stars, or other faint 

 objects dillicult to see, practical astronomers look for them 

 sideways, out of the corners of their eyes. The centre of 

 the retina is not so sensitive as the outer portions, which 

 are much less used. 



(To he continued.) 



A NATURALIST'S YEAR. 



By Grant Allen. 



VII.— WILLOW CATKIX.S. 



rilHE tall lithe sallows in the hedge beside the river are 

 X now all bursting out bravely into silvery catkins, 

 some of them still covered by the silky down of early youth, 

 and others yellow with the golden pollen of the maturer 

 flowering season. Among them the bees are busy already; 

 for you hardly ever see a willow catkin in full bloom 

 without a bevy of its attendant fertilising insects. Unlike 

 most other catkin-bearing trees, indeed, the willows depend 

 entirely for the due setting of their seeds upon winged 

 allies, and that is why their catkins stand so closely and 

 stiffly set upon the branches, instead of hanging out loose 

 and pendulous to the passing breeze, like those of the hazel, 

 the birch, and the alder. There are no more curious flowers 

 on earth than the little crowded blossoms of the catkin- 

 bearers generally, and none among these themselves more 

 singular and more highly developed than the willow kind. 

 Let us begin by looking at the actual existing catkins of 

 the sallow-bushes here, and then consider by what line of 

 development they have reached their present strange con- 

 dition. 



The first thing one notices about the willow catkins is 

 the fact that they consist of two kinds, growing apart from 

 one another on separate bushes. The one kind, containing 

 only female flowers, consists of silky-looking chisters ; and 

 when one pulls such a cluster to pieces, it turns out to be 

 made up of single scales, each enclosing one small spindle- 

 shaped ovary, from which are finally produced the familiar 

 tufted cottony seeds, that fly about so abundantly through 

 the air a little later in the season. Between this long 

 ovary and the central common stalk of the catkin stands a 

 little glistening, sticky gland — the nectary, in fact, whose 

 honeyed secretion attracts the buzzing bees that flit 

 so constantly about the flowering osier beds. The 

 other, or male kind, consists of long cylindrical branches ; 

 and these branches, when similarly taken to pieces, disclose 

 a number of other small scales, having two stamens con- 

 cealed within each. Between these stamens and the stalk 

 stands another sticky gland, just like that of the female 

 flowers, to complete the attraction for the fertilising bees. 

 In this very degenerate type of blossom, therefore, each 

 flower is reduced either to a single pistil or to a pair of 

 solitary stamens, covered only by a small concave bract or 

 scale. 



In order to understand how the willow flowers have 

 ever reached such a stage of degradation as this we must 

 trace the various downward steps by which their catkin- 

 bearing allies have gradually lost all semblance of a per- 

 fect pctaliferous blossom. The best point from which to 

 start on our investigation is given us by the family of the 

 elms and nettles, which are probably very close relations 

 of the catkin-bearing trees. These plants have small 

 fliowers with four sepals or calyx-pieces in each, no petals. 



and four stamens. But we can easily see that they are 

 descended from petal-bearing ancestors, because the 

 stamens are placed opposite the sepals, instead of alter- 

 nately with them ; and we know tliat in perfect flowers 

 each whorl or row of organs always alternates with the 

 whorls within and without it. In fact, wherever we find 

 abortive or monstrous petalless flowers, belonging to kinds 

 wliich usually produce petals, they have their stamens thus 

 opposite the calyx-pieces, instead of alternating with them. 

 Hence we are justified in concluding that the elms and 

 nettles, with their opposite stamens, are descended from 

 ancestors which had bright-coloured corollas ; but that the 

 corollas have been lost through disuse, and have now left 

 no token of their original presence, except this structural 

 peculiarity in the modem flowers. The elms, indeed, have 

 once more reverted to insectfertilisation, and so their 

 calyx has become bright pink in colour, and they also 

 still retain a pistil and a set of stamens in every 

 blossom — in other words, they are, as it is tech- 

 nically called, hermaphrodite. But the nettle?, which 

 are purely wind fertilised, have gone a step further 

 on the downward course, and have become uni- 

 sexual ; that is to say, each flower has either a pistil 

 alone, or stamens alone, as is also the case in our willows 

 here. Moreover, the grouping of the flowers of our com- 

 mon nettle in great loose bunches distinctly suggests the 

 first beginnings of the catkin form. Such a loose arrange- 

 ment of the male and female flowers in very mobile masses 

 is, of course, peculiarly appropriate for wind-fertilised 

 plants, as it enables the pollen to be shaken out easily by 

 every breath of wind, and conveyed with safety to the spot 

 where many sensitive surfaces are spread out at once for 

 its due reception. 



From the nettles to such early and regular catkin flowers 

 as those of the alder is not a long or serious step. Here 

 you see an alder branch in full liloom, with its long pendu- 

 lous male catkins hanging pensile to the wind in graceful 

 clusters. At first sight they look not unlike the male 

 willow in their arrangement ; but when you come to scan 

 them closely, you see that under each hollow bract or scale 

 there are not merely two stamens, as in our willow, but a 

 little cluster of three tiny florets, exactly like those of the 

 nettle, only on a smaller scale. Each floret contains four 

 calyx-pieces, with four stamens opposite them ; so that 

 here again we get indirect evidence of the suppression 

 of a former petal whorl. In the birch we find a 

 further stage of degradation. The scale, or bract, now 

 subserves all the purposes for which the calyx was origi- 

 nally intended ; and as each bract covers and protects three 

 distinct florets, there is now no special need for keeping the 

 calyx-pieces separate, or for maintaining the individuality 

 of the florets themselves. Accordingly, nature seems, to 

 some extent, to have lost count, so to speak ; there are still, 

 as a rule, twelve stamens (making up the original three 

 florets) under each bract ; and there are also still, as a rule, 

 twelve tiny sepals (making up the original three calyxes) ; 

 but sometimes the number falls to ten, nine, or eight, and the 

 sepals do not form a regular caixy, but lie about indiscrimi- 

 nately among the stamens. Even here, however, a sharp eye 

 can yet detect the surviving traces of three distinct florets. 

 Jn the hornbeam and hazel all trace of the calyx is lost ; 

 in other words, the tiny scales or sepals are now no longer 

 produced, and wo get only a lot of loose stamens included 

 \\ithin a covering l)ract. Yet the hornbeam keeps up a 

 very vague memory of the three primitive distinct florets, 

 for it has usually about twelve stamens under each bract ; 

 while in the hazel, even this last trace of the three-fold 

 arrangement has faded away, and the stamens are generally 

 reduced to about eight To the very end, however, the 



