2 



♦ KNOV\ALEDGE ♦ 



[Jdlt 6, 1883. 



a small pocket lens. The opening is, in fact, merely one 

 of the ordinary devices to prevent self-fertilisation ; it is 

 found in all the larger and handsomer British species, 

 which specially lay themselves out to attract in&oet visi- 

 tors, while it is wanting in the smaller and less conspicuous 

 varieties. Nor is this all, for in the little intermedixte 

 type, known as Epilobium roseum, the lobes sometimes 

 open t^hortly, and sometimes remain closed in a club-shaped 

 stigma : and I have found flowers in both states growing 

 together on the same plant. 



A rapid glance at the various forms assumed by British 

 representatives of this Protean genus will help better than 

 any amount of generalised talk to explain exactly how 

 closely the different groups resemble one another. Omit- 

 ting the naturalised French willow or rose-bay (which I do 

 not believe to be truly indigenous anywhere in Britain), we 

 have four fairly distinct central types of native epilobes, 

 as it seems to me, usually distributed into from seven to 

 eleven or more recognised species. Of these the largest, 

 handsomest, and, on the whole, most marked type, is the 

 great purple-red willow-herb (Epilobium liirsutuini), with a 

 bunch of very big and showy flowers. This is the great 

 insect-fertilised form, laying itself out most for attractive 

 display, and having the stigma deeply four-lobed. Below 

 it comes the so-called broad epilobe {E. ntontanum), differ- 

 ing from it, in good typical specimens, by its stalked 

 and rounded leaves, while those of the willow-herb 

 are longer, narrower, and more sessile on the stem, 

 as well as by its much smaller and pinker flowers, 

 still, however, with a four-lobed stigma. This is 

 a less luxuriant form of the same general type, growing in 

 dry situations, while the willow-herb affects moist ditches ; 

 and it shows some signs of incipient degeneration in its 

 short petals, sometimes twice as long as the calyx, but 

 often only of about the same length. These two forms are 

 universally admitted as distinct. But intermediate between 

 them come the specimens known as E. parvifloricin, some 

 of which closely resemble the willow-herb in leaves and 

 flowers, while others approach very near to E. montamun. 

 So far as my observation goes, in very wet situations, they 

 verge in the direction of the larger type, and in very dry 

 ones of the smaller. I may add that dwarfed specimens of 

 E. montanum, growing on the top of walls or other water- 

 less spots, carry the characteristics of the small type to 

 an extreme. But between these intermediates, even, it is 

 possiljle to find yet other intermediates, approximating 

 still more closely to the upper or lower forms. So much 

 for the two main large -flowered types with four-lobed 

 stigmas ; next as to the two main small-flowered types, 

 with a club-shaped style. One of these is a lowland form 

 {E. tetragonum) ; the other is a low, tufty mountain plant 

 (E. alpimim). But E. tetragonum, once more, which in 

 very typical specimens is known by its square stem, 

 marked with four decurrent lines, is linked to the four- 

 lobed species by the intermediates known as E. roseum, 

 some of M'hich have the style club-shaped, while others 

 have it slightly opened into four short lobes. This E. roseum 

 itself merges into E. montmuim, on the one hand, by the 

 small-flowered dry-soil specimens mentioned above, through 

 the shortly four-lobed form called E. lanceolatum, and into 

 the regular E. tetragonum, on the other hand, by specimens 

 with entire stigmas and marked decurrent lines. Other 

 small diflerences, which I need not note here, have made 

 some divergent plants of E. tetrngomi m be considered as 

 specifically distinct, under the name of E. virgatum. 

 Finally, the dwarf mountain type, E. alpinum, 

 when growing under exceptionally favourable circum- 

 stances, by the banks of Highland streams, assumes 

 longer and broader leaves, toothed on the edges, and 



recurs somewhat towards the lowland type, in which state 

 it is known as E. alsincefolium ; while in boggy places on 

 lower ground, it reverts towards the square-stemmed form, 

 and 1 ^ceives the name of E. jialustre. " It sometimes, also, 

 oc-nes very near the narrow-leaved forms of E. roseuia and 

 E. tetragonum," says Mr. Bentham, " but has the buds 

 much more nodding, and the decurrent lines on the stem 

 are either very faint or entirely wanting." For my own 

 part, I have seen intermediate specimens between almost 

 all these carefully-distinguished species or varieties which 

 it would be mere juggling, in my opinion, dogmatically to 

 refer to either or any type. But if ever the various inter- 

 mediate forms are weeded out by the action of natural 

 selection, there will probably remain four " good species," 

 as systematists call them : — E. hirsutum, E. montanum, 

 E. tetrago^ium, and E. aljjinmn. 



This is rather a long, dull, and technical exposition, I 

 am aware ; but it will serve better than anything else to 

 show how impossible is the attempt to draw definite lines 

 in such very mixed and elastic families. Is or are these 

 families by any means exceptional. To anybody who 

 wishes for an insight into the variety and modifiability of 

 plant forms even within the limits of a single small 

 country, the careful comparative study of the following 

 common English genera and species may be confidently 

 recommended : the reses [Rosa canina, R. arvensis, kc. 

 Mr. Bentham makes five species, and Mr. Babington, 

 seventeen) ; the brambles {Rubus, from four to forty-five 

 species recognised) ; the potentillas (including TormentiUa 

 and the strawberry) ; the St. John's worts (Hypericum) ; 

 the sow-thistles [Sonchus) and the hawkweeds {Hieracium). 

 How any one can rise from a close comparison of the 

 endless forms of hawkweed alone, a believer in strictly 

 defined natural species, is one of those psychic mysteries 

 only adequately to be explained by Mr. Bishop, or by 

 Messrs. Maskelyne & Cooke. 



THE CHEMISTRY OF COOKERY. 



XIII. 



By W. Mattieu Williams. 



THE process of frying follows next in natural order to 

 those of roasting and grilling. A little reflection will 

 show that in frying the heat is not communicated to the 

 food by radiation from a heated surface at some distance, 

 but by direct contact with the heating medium, which is 

 the hot fat commonly, but erroneously, described as 

 " boiling fat." 



As these papers are intended for intelligent readers who 

 desire to understand the philosophy of the common pro- 

 cesses of cookery, so far as they are understandable, this 

 fallacy concerning boiling fat should be pushed aside at 

 once. 



Generally speaking, ordinary animal fats are not boilable 

 under the pressure of our atmosjjhere (one of the con- 

 stituent fatty acids of butter, butyric acid, is an exception; 

 it boils at 2>\i° Fah.). Before their boiling-point, i.e., the 

 temperature at which they pass completely into the state 

 of vapour, is reached, their constituents are more or less 

 dissociated or separated by the repulsive agency of the 

 heat, new compounds being in many cases formed by re- 

 combinations of their elements. 



When water is heated to 212° it is converted completely 

 into a gas which gas returns to the fluid state without any 

 loss on cooling below 212°. In like manner if we raise 

 an essential oil, such as turpentine, to 320°, or oil of 

 peppermint to 340°, or orange peel oil to 34.5°, or patchouli 

 to 489°, and other such oils to various other temperatures, 



