210 



KNOWLEDGE. 



[September 1, 1899. 



each column, according to the five sunspot maximum 

 years dealt with. 



Now, we may here be struck by the fact that the hottest 

 summers cluster in the earlier years of the series, that is 

 to say, about the time oi nrouth and maxima of the spots. 



Suppose we draw a line at twenty-four, and call every 

 summer season with more than twenty-four of those days 

 very hot. We find twelve such summers, and ten of these 

 are in the first five columns (3, 2, 1, maxima, 1). In 

 the years after maxima are passed, the dots cluster nearer 

 the average line ; the summer heat, in fact, is moderated. 

 Coming to the end of the series, i.e., near sunspot minima, 

 we find (6, 7) a large preponderance of cool summer 

 seasons. 



The same general view seems to be borne out by the 

 diagrams B and C ; in the former of which seventy degrees 

 is taken as the thermometric limit, instead of eighty 

 degrees ; while in the latter we consider the mean tem- 

 perature of May to August. 



Thus, in the case of B, we find tm summers which, 

 having more than ninety days with thermometer up to or 

 over seventy degrees, might be called veri/ hot. Of these, 

 as many as nine are in the first five columns, and o7ie only 

 in the remaining six. True, that one (1865) is the hottest 

 summer of the whole, from the present point of view (with 

 one hundred and thirty- two days), and so presents a curious 

 anomaly. The details of C may be worked out, similarly, 

 by the reader. 



Comparing these three diagrams, we may further notice, 

 I think, in years following maxima, a constriction (so to 

 speak) of the space covered with marks, from helow as well 

 as from above, towards the average line. In other words, 

 the summers in those years apparently tend to be less 

 extreme in coolness as well as in heat. (Consider the 

 columns 2, 3, 4, after maxima, in B and C especially.) 

 ■ Reverting to A, it may be useful to specify those twelve 

 hot summers (as measured) ; they are 1846, 1847, 1857, 

 1858, 1859, 1868, 1870, 1876, 1884, 1887, 1893, 1895. 

 The hottest is 1868 (forty days). The seasons 1876 and 

 1887 form the exceptions to the rule. 



Now, the above mode of classification (starting from 

 maximum sunspot years), though adequate as far as it 

 goes, does not quite exhaust the summers since 1841, and 

 slightly fails in regard to minima ; it is well, therefore, 

 to supplement it by another, in which we start from 

 minimum sunspot years, counting forwards and backwards. 



In the fourth diagram, D, we are enabled, taking the 

 same measure of summer heat as in A, to compare the 

 five years endiiuj with sunspot minima, with the five years 

 folloxcing those minima. The result is very much the 

 same — few very hot summers in the former group, and 

 in the latter part of this period of spot decline, a large 

 preponderance of cool summers. But once the minima 

 are passed, and the spots are growing, a number of very 

 hot summers. 



It is in accordance with these results that we generally 

 find, in regard to summer rainfall at Greenwich, a pre- 

 ponderance of wet summers in the five years ending with 

 sunspot minima, and a preponderance of dry summers in 

 the five years following minima. 



The practical view, then, to which these inquiries point 

 (and it is offered without dogmatism as to sunspot 

 influence, or affirmation of an unvarying law), is that we 

 are at present at a time when cool summers are to be 

 looked for rather than hot ones, and that only after the next 



sunspot minimum is passed (say 1901, but not certainly), 

 need we prepare for recurrent times of thorough and 

 persistent broiling ! ^' 



THE STORY OF THE ORCHIDS.-II. 



By the Rev. Alex. S. Wilson, m.a., b.sc 



THE butterfly orchises, Habenaria bifolia and H. 

 (Iilorantha, are also fragrant; they have greenish- 

 white flowers with very long spurs. The greater 

 butterfly orchid is perhaps the deepest of all our 

 native flowers ; its nectary sometimes attains a 

 length of one and a half inches. Its fertilizers are 

 sphinx moths, which alone of all our native insects 

 have a proboscis sufficiently long to remove the 

 nectar. The pale colour of the flower and the absence of 

 markings are also indicative of nocturnal or crepuscular 

 visitors. The rostellum differs from that of the orchids 

 already considered ; the two viscid disks are separate and 

 placed at the sides of the entrance to the spur. In this 

 position they may adhere to the moth's proboscis, or to 

 the sides of its head ; not unfrequently they are observed 

 attached to its eyes. 



* That is, 1848, 1860, 1870. 1884, 1894. As there are only four 

 summers since 1894, 1 have made use of the summers 1842, 1843, 

 1844, being the fifth, sixth, and seventh after sunspot max., 1837. 



Fi&. 4. — I, Hahenaria :, 2, Epipactis ; 3, Lisfera, 



In the fly orchis, Ophrys muscifera, the rostellum is also 

 divided, each disk being lodged in a separate cup. The 

 front view of this orchid presents a remarkable likeness to 

 a fly ; from this circumstance it gets its name. But the 

 name is appropriate for another reason, for the flower 

 seems to be adapted to carrion flies. The labellum is not 

 spurred, but secretes minute drops of fluid which are 

 gathered by the visitors ; the latter as they move upwards 

 come against the rostellum and remove one or both 

 poUinia. Two black shining bodies at the base of the lip 

 are believed by Miiller to be pseudo-nectaries intended to 

 deceive certain classes of visitors. 



The twayblade, Listera ovata, is another orchid frequently 

 met with in woods and pastures. It has yellowish -green 

 flowers, which, as in the previous case, curiously resemble 

 one of its most frequent visitors — a beetle, Grammoptera 

 levis. The bilobed labellum hangs down in front ; it has 

 no spur, but honey is secreted in a longitudinal furrow. 

 Besides the above-mentioned beetle the twayblade is visited 

 by ichneumon flies. As they creep upwards, licking the 

 honey in the groove, the rostellum is touched and a drop 

 of fluid exudes which instantly hardens and cements the 

 pollinia to the head of the visitor. At the same time the 

 scale-like rostellum bends downwards and protects the 

 stigma from self-fertilization ; later on the scale becomes 

 again elevated so that the stigma is once more exposed. 

 The bird's-nest orchid is fertilized in an analogous manner. 



Another of these orchids with greenish-yellow, not 

 particularly attractive, flowers is the Helleborine, Epipactis 



* Up to 19th August, there have been this season, I think, sixteen 

 of these very hot days, eighty degrees and over. 



