58 



HARDWICKE'S SCIENCE-GOSSIP. 



obtaining and manufacturing nourishment for their 

 support ? 



I have shown in a former paper, how a root of a 

 tree can grow and form layers of woody matter after 

 being deprived of any visible means of obtaining 

 nourishment. We have observed how the knot is 

 formed. Once deprived of all visible means of ob- 

 taining matter by leaves of its own, it is fed by the 

 bark, and will live and increase yearly in size, 

 adding additional layers of woody matter, hence its 

 growth, and will live and increase in size in this 

 manner as long as the bark supplies it with matter. 

 These growths are many times irregular and curiously 

 twisted. The layers of woody matter when cut up 

 and polished, present a beautiful surface. I have 

 examined several hundreds of these knots of all ages 

 and stages of growth. I have found at least the 

 half of them have no connection inwardly, they have 

 lost connection (see Fig. 43). They are seated between 

 the bark, and are most easily knocked off the tree. At 

 this stage any one can take a knife and pick out the 

 inner bark which is between the knot and the layer 

 of wood, and see this for himself. 



Among the various forms to be found, these (see 

 Figs. 34, 37, 40) are the forms most likely to live 

 longest. These (38, 39), from their neck-hke con- 

 nection, are more likely to die off earlier, because 

 the sap, when descending down the stem, does not 

 circulate so freely around the knot as these others 

 which have a closer and better connection. 



It must not be supposed that all knots lose the 

 shoots. I have found several instances (see Fig. 41), 

 where both branch and knot continued growing. In 

 time, the knot would be enclosed in the growth of 

 the branch. 



On some trees, for instance, the Cryplomeria, I 

 have found knots of woody matter, not caused in the 

 same way as these described, but due to branches 

 which impede the sap and cause the bark to rise in 

 folds, so to speak, which form knots of woody 

 matter at such parts where there is a stagnation of 

 sap and impediment to the free enlargement of the 

 bark. 



The Scotch pines, both those of a moderate and 

 those of a larger size, are liable to curious knots. 

 I had often wished for some trees to be cut so that I 

 could get them dissected and examined. Latterly, 

 by the effects of the great gales, this opportunity did 

 come. I found that these growths (Fig. 30) were due 

 to broken and pruned branches, particularly if two or 

 three were quite close together. This caused a stagna- 

 tion and irregular deposit of woody matter, at those 

 parts which caused these curious growths. I have 

 found some very large and remarkable examples of 

 these growths. 



The larch is not very guilty of such growths. I 

 have found many instances of them. After inves- 

 tigating them, I found them due to broken and 

 to pruned branches, which caused an irregular 



growth of the bark overlapping, so to speak, resulting 

 in an irregular growth of the bark. The arrows in 

 each figure, mark the course of the ascending and 

 descending sap, 



Robert Coupar. 

 Ashjord Castle, co. Gahvay. 



COLOUR DEVELOPMENT IN LEAVES AND 

 FLOWERS. 



WITH regard to my quotations from Hermann 

 Miiller as to the visits of bees to the poppy 

 and two periwinkles, Mr. Bulman cautions me that the 

 absolute number of visits, not the number of species 

 observed on the flowers, is the important point. This, 

 of course, is quite true, but there is nothing to show 

 that in the cases mentioned, the distribution of the 

 total number of visits was not roughly proportional 

 to the number of species ; indeed, Muller's remarks 

 seem to indicate that this was so. Mr. Bulman's own 

 observations on the comparative number of insect 

 visits to the chickweed and Veronica Biixbaicniii, 

 growing in the same garden, are interesting, and with 

 his other observations certainly seem to tell against 

 the theory. I do not, however, feel justified in giving 

 up the theory of the production of blue flowers by the 

 selective action of bees merely on the strength of 

 these experiments, which are possibly capable of being 

 explained away on a further knowledge of the con- 

 ditions. With regard to Sir John Lubbock's experi- 

 ments, the interpretation seems plain enough, that 

 bees, from whatever cause, prefer blue to other 

 colours, and as to the inconsistency with " observed 

 facts," I must say that this is, in my opinion, far too 

 comprehensive a phrase to apply to a difference of 

 result arrived at by Mr. Bulman. At least, I hope, 

 it is clear that the experiments upon which Sir John's 

 opinion is founded are not a matter of "teaching" 

 bees " to take honey off different colours" ! 



Mr. Bulman says that "it is hardly possible to be 

 otherwise than dogmatic " upon questions touching 

 " advanced " flowers. This may be so, but he imme- 

 diately afterwards goes on to speak of the ' ' lowly mark 

 of symmetry," tic, and as he has ventured on to this 

 dangerous ground, I must, in justice to the theory 

 in question, follow him. To take first, then, his 

 instances of highly developed flowers which are not 

 blue. Mr. Grant Allen considers that the Orchid- 

 acea; "have mostly got beyond the monochromatic 

 stage altogether." As to the Boraginacece, I cannot 

 consider them at all a primitive type, though they do 

 possess "the lowly mark of symmetry." Their 

 honey-concealing tubes, the various appendages to 

 the corolla for excluding " unbidden guests," and 

 other contrivances for • ensuring cross-fertilisation, 

 surely mark thern out as highly developed. These 

 are among the characteristics usually considered to 

 constitute an organism "advanced" — increasing 



