KNOWLEDGE. 



July, 1911. 



which also we find higher adaptation to insect visits. In the 

 Sympetalae, ^ygoniorphy occurs in nearly lifty per cent, of 

 the species, as against barely fifteen per cent, in the case of 

 .■\rchichlainydeae. .■\ggregation into dense inflorescences is 

 the chief feature of the Compositae. an order which comprises 

 over twenty-seven per cent, of the total number of Sympetalae, 

 and. indeed, over ten per cent, of the sum total of flowering 

 plant species. 



The author then gives a summary of the various cohorts and 

 orders of Archichlamydeao. tracing the evolutionarv tendencies 

 through this series. Taking first the essential organs — the 

 androecium and the pistil — the general tendency to econonw 

 in production of parts is emphasized, and the question is raised, 

 ■■ To what extent has the working of the principle of economy 

 been accompanied by the compensatory tendency of progressive 

 adaptation to insect visits ? " In this connection attention is 

 called to the absence in Archichlamydeae of a strong tendency 

 io floral angregation. such as we find realized and e.xpressed 

 in the Compositae among the Sympetalae. Dense 

 inflorescences occur in isolated cases throughout the Low^er 

 Dicotyledons, but aggregation does not characterize any 

 relatively large group except one, the Umbelliflorae. in whicla 

 the inflorescence unit is typically a close umbel. The general 

 tendency to zygoiuorphy goes hand in hand with progressive 

 aggregation, the outer florets tending to bilateral synnnetry 

 and especially to increased development of the corolla on the 

 outer side. In the Lower Dicotyledons, however, the 

 zygomorphy of relatively large and solitary flowers occurs 

 only in isolated cases here and there, but does not form a 

 critical character of any great group comparable with the 

 Personales and Lamiales in the Sympetalae. .Again, fusion 

 plays a very inconspicuous part among the Lower 

 Dicotyledons : a corolla tube is, of course, absent, though 

 some specialised " apetalous " groups (e.g., Proteales) have 

 flowers with a typically gamophyllous perianth. In the 

 Geraniales and the Papilionaceae, however, a tube is formed 

 by the cohesion of the filaments of the stamens. But a far 

 more extensive tube-forming tendency in the .-Xrchichlamydeae 

 is that produced by means of the progressive hollowing of the 

 flower receptacle. 



In the higher families of .Archichlamydeae, from the 

 Buttercup cohort (RanalesI upwards, we find various stages 

 of pcrigyny. leading to cpigyny. The inferior position of the 

 ovary may. however, have been produced in descent in other 

 ways than by a gradualU- increasing degree of perig\ny, 

 though this method alone has left any continuous trace among 

 existing plants. Whatever the evolutionary methods of its 

 production may have been, the inferior position of the ovary 

 conduces to economy in production, since in this way 

 receptacular tissue can be pressed into the service of ovule 

 protection and fruit formation. At any rate, it is certain that 

 these floral types, such as the Compositae. which are 

 admittedly in the \an of evolutionary .advance, invariably 

 have an inferior ovary, w^hile no epigynons flower can be 

 called unquestionably primitive. 



In attempting to find some phxletic connection between the 

 .Archichlamydeae and the Sympetalae. one can hardly suppose 

 that the cohesion of the petals, which is the sole .essential 

 difterence between the two groups, followed upon one line of 

 evolution, nor that sympetaly originated at one point only of 

 any particular line. That is. there is a prima facie presump- 

 tion that the Sympetalae are polyphyletic in origin, yet even if 

 the series is composite it is also synthetic, in so far that its 

 component members may be connected on the lines of certain 

 well-marked progressive tendencies to biological advance. 

 These tendencies do not difier fundamentally in any way from 

 the tendencies seen at work in the .Archichlamydeae. but they 

 do differ in degree, as might be expected from the relatively 

 high biological organisation of the group. The tendency to 

 reduction of parts has already reached an advanced stage of 

 realisation, and it plays a secondary role in the Sympetalae. in 

 which the prominent factor is that of progressive .adaptation 

 to insect visits. 



In the Sympetalae. tlie (endency to economy in production 

 is expressed by («! reduction of stamens to a number less 

 than that of the corolla segments, following upon zygomorphy : 



<h) progressive reduction of the calyx to a pappus or to zero; 

 k") reduction of the carpels to two. the loculi to one. and the 

 ovules to one. 



The tendency in Sympetalae to progressive adaptation to 

 insect visits is shown in (al the general introduction to 

 zygomorphy in Tubiflorae on the one hand and in .Aggregatae 

 and Campanulatae on the other, representing the two types of 

 zygomorphy shown in .Archichlamydeae; (b) the pollen 

 presentation mechanism, a general feature of the Campanulatae 

 and especially of the Compositae order. 



CHEMISTRY. 



By C. .AiNSWORTH Mitchell. B..A. lOxon.), F.I.C. 



MF.ASLKEMEXT OF POLLUTION OF THE AT.MOS- 

 PHFKF. — Hitherto, the only chemical means of determining 

 the degree of pollution of the atmosphere has been to estimate 

 the proportion of carbon dioxide, and draw a deduction from 

 the amount by which it exceeded the normal quantity. This 

 method could only be regarded as giving approximate values, 

 since carbon dioxide is one of the products of combustion as 

 well as of respiration, and its estimation aftbrded an indication 

 of the degree of ventilation rather than that of the vitiation of 

 the air, which is due to the condensation of animal products 

 of excretion. 



.An ingenious method of measuring these condensed products 

 has been devised by MM, Henriet and Bouyssy [Coiuptcs 

 Rendiis. 1911, CLIL, IISOI. It is based upon the fact that 

 pure air always possesses oxidising properties, whereas 

 excretory products are all reducing substances, and thus an 

 estimation of the reducing property of a measured volume of 

 the air is strictly proportional to the degree of pollution. The 

 foreign bodies of organic origin to which pollution is due, 

 are condensable simultaneously with the moisture in the 

 atmosphere, and the method, therefore, consists of three 

 steps: — (1) Condensation of a gi\en volume of water from the 

 atmosphere by means of a freezing mixture; (21 Calculation 

 of the quantitN' of air corresponding to the amount of water 

 condensed: and 13) Estimation of the amount of potassium 

 permanganate solution reduced by the organic impurities in an 

 aliiiuot part of the condensed moisture. The results are 

 calculated into the corresponding quantity of oxygen, and 

 expressed in milligrannnes per one hundred cubic metres of air. 



By this method the relatively pure air in the Montsouris 

 Park, in Paris, gave a value of one, whereas the air in the 

 vicinity of the H6tel-de-Ville had a value of ten. .A room in 

 which several rabbits and monkeys had been kept gave the 

 value thirteen, and the same figure was obtained with the air 

 in a printing works. The figure fourteen was given by the air 

 in a badly-ventilated office, and seventeen by that in a dark 

 passage, while a dressmakers' work-room in which twenty 

 people had been working all day with the windows closed gave 

 the extremely high value of twenty-one. It seems probable 

 that this method will prove a very valuable aid to the hygienic 

 investigation of the air in closed places, such as submarine 

 vessels. 



BACTERIOLOGICAL STUDY OF HONEY.— An 

 interesting investigation of the species of bacteria present 

 within the cells of the honeycomb derived from different 

 parts of France, has been published by MM. Sartory and E. 

 Moreau \.\niiales des Falsifications. 191 1. I\', 259). .Among 

 the bacteria of common occurrence in the air. w-hich had been 

 introduced by the bees into the cells, were the following 

 species: — B. siibfilis. B. nicgateriiini, Sarciiia Intea, B. 

 aeropliiliis. and Sfaphylococcits pyogenes. There were 

 also several mould-fungi and yeasts, including Peniciiliuni 

 glaiicuni, Saccharoinyccs ccrevisiae. Mucor raceinosiis and 

 .Aspergillics gracilis. .Among the bacteria was a golden 

 yellow species, which secreted a bright yellow pigment soluble 

 in absolute alcohol. It had a strong liquefying action upon 

 gelatin, and differed in its biochemical characteristics from 

 several very similar yellow bacteria. It appeared to be 

 intermediate in its properties between the Bacillus liitcKs of 



