a Cor id wi iii 24 Department of Agriculture, Madras. Vol. IV.—Bulletin No. 69. BEES AND THE FERTILIZATION OF COFFEE BY T. BAINBRIGGE FLETCHER, RB.N., F.us., F.E.S., F.2.8., —_— P Imperial Entomologist, Pusa-— MADRAS: PRINTED BY THE SUPERINTENDENT, GOVERNMENT PRESS, \8 annas.] 1915. (3 penee.} Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924003581646 Department of Agriculture, Madras. Vol. IV.—Bulletin No. 69. BEES AND THE FERTILIZATION OF COFFEE BY T. BAINBRIGGE FLETCHER, R.N., F.us., ¥.E.S., F.Z.8., Imperial Entomologist, Pusa. MADRAS: PRINTED BY THE SUPERINTENDENT, GOVERNMENT PRESS. 1915. : BEES AND THE FERTILIZATION OF COFFEE. The question of the value of bees in coffee districts was raised at the annual meeting in 1911 of the United Planters’ Association of Southern India, with particular reference to the effect on outturn of crop caused by the regular destruction of these bees in certain districts (notably the Shevaroys), such destruction being alleged to cause the very marked diminution in the number of bees which is perceptible in such districts. After some cliscussion, the following resolution was proposed by the delegate for the Shevaroys and seconded by the Nilgiri delegate and passed :— ** That Government be asked through the United Planters’ Associ- ation of Southern India to pass some ruies prohibiting the destruction of bees in planting districts.” 2. The object of this resolution was to prevent or diminish the destruction of the bees themselves but not necessarily to place any impediments on the collection of honey or bees-wax. What was asked for was that Government should enforce rules in the coffee-planting districts to ensure that the bees themselves were not destroyed when their produce was taken. It was considered by the planters that it should be possible to take honey and wax without destroying the bees and that the present method of burning the bees off the combs was adopted simply as entailing the minimum amount of trouble. 3. In reply to this.request, instructions were issued by Government that the matter should be investigated by the Entomologist. (See G.O. Mis. No. 2062, dated 11th July 1912, and addressed to the Secretary of the United Planters’ Association.) 4. In order to determine as far as possible the actual importance of bees in the pollination of coffee, it was necessary to carry out field experiments to test this point. It was also requisite to find out definitely what species of bees were concerned, to what extent they were actually destroyed and how far such destruction had really effected any appreciable diminution in their numbers and, if such diminution seemed the effect of robbing the nests for honey and wax, how this destruction of bees could be minimised. 5. Unfortunately I took over the duty of the newly-created post of Government Entomologist in Madras in April 1912, too late to make any such experiments in that year, as inquiry showed that the coffee had already blossomed in all the coffee districts, and it was necessary to defer the experiments until the next year. In order to gain a preliminary idea of the districts concerned, to find out what bees occurred there and to make local arrangements for the conduct of experiments during the ersuing coffee-flowering season, I paid a visit in October-November 1912 to the Shevaroys, the Bababudin hills, and to Coorg, and thereafter submitted the following tour report :— I.—BRepost or ta GovERNMENT ENTOMOLOGIST ON A TOUK IN THE SHEVAROYS. “T left head-quarters on the evening of 13th October, arriving at Salem the same night and proceeded next morning to Yercand. 1-4 4 Here I examined some coffee and in the evening went on to Hopeville estate. “On 15th I visited Kadiar rocks, a high rocky cliff to which the wild bees (Apis dorsata) suspend their combs, whence a large quantity of honey and wax is collected yearly. In the evening I visited Arcadia estate, where a few colonies of Apis indica have been kept in English hives, but unfortunately these had all swarmed and. deserted the hives just prior to my visit. Investigation of the combs showed that wax moth (Galleria melionelia) had effected an entrance, (See Agricaltural Journal of India, October 1911.) Fie@, 1.—Where Apis dorsata builds its comb. ‘I'he Kadiar rocks in the Shevaroy hills. The bees build below the overhanging rock shown l@:e and the honey-gatherers are lowered over the precipice by rores. Fig. 2.—-Diagrammatic sketch of a bar-frame beehive, showing construction. “On 16th I observed flower-haunting insects and did some general collecting, visiting Swiuton estate and Scotforth estate, to see local conditions and make general observations on insects. “On 17th I proceeded to Vallakadai Peak estate, collecting and observing in the intervening country. “On 18th I observed flower-haunting imsects, gathered intor- mation regarding bees, and saw and advised on various pests of coffee, orchard and garden crops. It was a wet misty day very unfavourable for collecting or observing insects. At about 6-30 p.m. a sharp earth- quake shock, lasting about one second, was felt here. “On 19th, as the north-east monsoon seemed to have set in and it was misty and raining hard, I left for Yercaud, reaching there the same afternoon. “On 20th (Sunday) I halted at Yercaud. It was a misty, wet, cold day and few insects could be found, but I was able to obtain several interesting species. “On 2lst it was still wet and unfavourable and I left in the afternoon for Salem and arrived back at Coimbatore on the morning of 22nd October. , “ Three species of true honey bees (Apis dorsata, indica and florea) appear to occar in the Shevaroys (I omit consideration ofa fourth re- ported species, which is evidently a Melipona). Of these I was unable 6 to come across a single specimen of Apis florea, and neither of the other two species was at all common. The absence of bees in these hills at the time of my visit was indeed remarkable. During the last three or four days the weather was, it is true, extremely unfavourable, yet even this would hardly account for the almost total absence of bees, and I am inclined to suppose that Apis dorsata (and perhaps Apis indica also) migrate from these hills into the plains during about July- November, This would be in accordance with the habits of Apis dorsata elsewhere (notably at Sabour,in Bihar, as Mr. Wvodhouse has informed me) but further observations are desirable, as the point is of course not only interesting from a bionomic viewpoint but also of practical importance in considering the damage to these bees caused by the present methods of collecting honey and bees-wax. ‘“ The right to collect these is leased out by the Forest Department, usually to Malayalis living in adjacent villages. The bees (by which I refer to the large rock-bee, dps dorsata) suspend their large single combs from the lower surface of inwardly sloping rocks which occur usually as high cliffs rising for 200 or 300 feet. As these combs are usually affixed to overbanging rocks near the top of such a cliff, the honey-gatherers generally climb down by a rope ladder fastened from the top. Occasionally the ladders break, or the gatherers are so badly stung that they lose their hold and drop, being dashed to pieces on the rocks below. ‘They are provided with a blanket and a torch, with which they burn the bees off the combs, which are then cut and col- lected. The destruction of bees by this process, which takes place in May-June, is enormous as 1 am informed by all observers. At the same time, Iam doubtful whether the honey and wax could be collected without the destruction of the bees, which would otherwise pursue the plunderers of their stores and exact full vengeance on their disturbers. ‘“‘ Apis dorsata seems to be a fierce and intractable bee whose habit of building a single large comb unfits it for domestication. Its migratory habits also, if these are established, form an obstacle from an apicultural viewpoint. Fic. 3.—Apis dorsata, worker. ‘The small outline figure shows the natural size. 7 “ Apis indica, a bee rather similar to, but smaller than, the English honey-bee, builds its nest as a rule in hollow trees where it forms several parallel combs. It can be hived, although it is very subject to attack by wax-moth whereupon it usually deserts the nest and swarms off elsewhere, so that it is rare to induce a colony to remain in one hive for more than a few months. ‘The ordinary English bar-frame hive and foundation comb, being made to suit the larger European bees, are hardly suitable for Apis indica, whose combs are smaller, averag- ing about 35 cells in 6 inches, whereas European combs run to about 27. This difficulty may be overcome by the use of specially made foundations and I am writing to Colombo to make inquiries regarding the availability of such for Indian bee-keepers, as I understand eX Fanndatied “iachiad “ior thatthe Ceylon Agricultural Society Apis indica is now available at has recently ebtemice a found. Pusa and foundation can be ®10> machine * specially built to made if wax ia supplied. suit Apis indica. This bee offers the most likely subject for trial in T.B.F.—15-6-14. domestication amongst the native Indian bees, although, as remarked above, it is not altogether satisfac- tory. It is at times rather a vicious species, resenting interference and . manipulation, and that also is a point against asking coffee planters to endeavour to domesticate it. I presume that its honey and wax are covered by the term “forest products ”; but I am unaware whether collection of products of this variety is included under the licenses granted by the Forest Department. In any case, it would prove most difficult to enforce protection for this bee, as it builds almost anywhere in the jungle wherever a suitable hollow tree is to be found and is not confined to special localities as is the rock bee. Fic, 4.—Apis indica, worker. The small outline figure shows the natural nize. ‘ Apis florea occurs in the Shevaroys by all accounts, although I saw none at all, It buildsa small single comb rarely more than 6 inches broad by 12 deep (often smaller) and is quite unsuitable for domestication. It usually hangs its comb from a branch of a bush where it is easily found and taken and this habit renders it (like Apis indica) very difficult to protect by any legislation. It is, though small, a most active little worker and must doa great deal of good in the pollination of flowers. Its store of honey is so small that itis not likely to be taken for commercial purposes but would doubtless be secured for present eating by jungle tribes. Fic. 5.—Apis florea, worker. The small outline figure shows the natural size. “ Although very few becs were to be seen at flowers by day, in the evening numerous moths, notably Nephele hespera. and Protoparce convolvuli, were attracted, whilst in the daytime Macroglossum spp. also darted around the blossoms. It will of course be necessary to determine by actual experiment and observation, when the coffee flowers next April, what proportion of flowers are actnally fertilized by bees, but I rather expect to find that other insects (such as the sphingids named above) will be found to have a large share in fertilizing them.” II.—Report of THe Government ENTOMOLOGIST ON A TOUR IN THE Bazasuoin District or MysoRE Strate. “T left Coimbatore on the evening of 30th October and next morn- ing reached Bangalore where I met Mr. G. N. Frattini, the Planting Expert’s Assistant for Mysore, and we reached Kadur the same after- noon. Owing to difficulty in obtaining transport, we started from Kadur early next morning and reached Chickmaglur the same afternoon. Next day (2nd November) we went on to the Bababndin hills, where 1 stayed until }3th November, when I left for Chickmaglur and Kadur and arrived at Mysore on the afternoon of 14th November, Here again there were transport difficulties aud it was not until noon on 15th that I was able to start for Sidapur which was finally reached at 3am. next morning. From Sidapur I went to Fairlands estate on 16th November, on 17th on to Margalli, and on 18th from Margalli to Mercara, where I halted for three days. From Mercara my intention was to visit Dunkeld estate, where considerable damage to cardamoms by a scolytid beetle has been reported ; but although the distance from Mercara was only some 20 miles, no means of transport were forth- coming and J was reluctantly compelled to abandon this visit. I left Mercara on 22nd November, stayed that night at Santikoppa, and proceeded on to Mysore next day, arriving back at head-quarters on the morning of 25th November. “The main object of this tour was to make a preliminary survey of the localities visited with a special view to collecting information 9 regarding the occurrence of honey-bees and other flower-frequenting insects, with regard to their influence on the pollination of coffee. Incidentally, of course, collections of economic and other insects were made and various interesting habits, ete., noted, but this report is not the place for such details. As the conditions are rather different the Bababudin hills and Coorg may be considered separately. The Bababudin Hills. “ Three species of true honey-bee occur here— (i) The large rock hee (Apis dorsata). (ii) The Indian bee (dds indica). (iii) The small bee (Apis florea). “ The large rock bee builds regularly at a height of about 5,800 feet ou the eastern face of the hills, recourse being had to one particular vertical cliff where there is a fair amount of shelter by overhanging rocks. This colony is not a large one, comprising perhaps thirty to forty combs on the average. | Fairly definite evidence was obtainable that these bees desert their ‘sombs on the onset of the south-west monsoon (about June) and return again with the north-east monsoon (November-December). Beyond the general upinion that they fly off in an easterly direction no information was obtained regarding the locality in which they spend the intervening period. Possibly it 1s somewhere in the vicinity of Kadur. The swarms which return to the hills at the end of the year seem to be small ones and perhaps the number of individuals on a comb is greatest at the end of the hot weather (April-May). Neither honey nor wax appears to be collected systematically whilst the bees are in the hills. The cliff on which they build is situated well above the limits of the coffee estates on this side of the hills and the bees, so far as I could ascertain, are not interfered with at all. The reason for their migration from the hills in June is probably to be found in climatic causes, the excessive rain in the hills during the south-west monsoon hindering them from foraging for food. «The Indian bee occurs fairly commonly, building asa rule in hollow trees. It is noteworthy that in these hills the local race of this bee is much darker than in the plains and this seems to be the case generally in the montane districts of Southern India. The nests are robhed when found and swarms are sometimes taken and kept in earthen pots although I only heard of one case of even this simple form of apiculture. “The small bee was not common at the time of my visit and I was only able to find one small colony, Possibly it migrates in the same way as the rock bee. This bee is regarded as stingless by the local inhabitants and the nests are commonly robbed for the honey which is usually squeezed out and eaten on the spot. “In addition to the true honey-bees, mention may be made of a large tawny-yellow carpenter bee (Xylocopu rufescens) which was found fairly commonly at flowers after dark. None were to be seenin the day time but they began to appear in the evening after sunset and flew freely around flowers long after night had fallen. This habit on the part of a bee was quite new to me but, on looking up the 2 10 literature, I find that Colonel Bingham recorded its crepuscular flight: (Journal of the Bombay Natural History Society, Vol. VIII, page 388) ; he also found the nest, which consisted of a series of round cells hollowed out in the end of a decaying log. This bee does not seem to have been recorded from Southern India before, although its peculiar nocturnal habits and comparative abundance (in some localities in these hills at least) may possess some importance so far as concerns the pollination of coffee. “Special attention was paid to the insects frequenting flowers. Those found in the daytime call for little comment, including the butterflies, bees, flies, etc,, usually found. ‘l'owards sunset, however, quite a different fauna put in an appearance, the first arrival at the flowers being hesperid (skipper), butterflies and humming-bird hawk- moths (Macroglossum spp.). As twilight sets in, these gave way to numerous other sphingids (Hippotion boerhavie, Deilephitahypothous, Theretra nessus, Herse convoluuli, Nephele didyma, etc-), which flew actively from flower to flower accompanied by numerous smaller- moths, especially various species of Plusia and others of slower flight. The number of flowers visited in an evening by one of these larger hawk-moths must be very large and it is notable that flowers with very deep nectaries are probably adapted for pollination by means ot scme of these moths whose tongues in some species (¢.g., H. convolvult) are much longer than their whole bodies. I expect we shall find, when experiments are made in the coffee-flowering season, that crepuscular and nocturnal insects such as these play a large part in the fertilisation . vf the blossom, and it may then be considered whether they cannot be encouraged by the planting of attractive bushes such as Madras Thorn (Durania). : “Nests of a small wasp, called locally ‘ Berna’ or ‘ Flying Ant’ are common in this district, being hung from trees. Though often: burnt out as a nuisance, these wasps are probably beneficial by feeding on caterpillars, etc., and rarely attack unless disturbed.” IIL.—Report or roe Government Enromo.ogist oN A TOUR In Coora. ‘My visit to Coorg was brief and hurried and gave little oppor- tunity for detailed investigations. The three species of honey-bee, mentioned. above occur. Apis dorsata (the rock bee) in this district seems to build on trees rather than on rocks, returning year after year, to the same tree, or group of trees, upwards of a hundred. combs being sometimes suspended from the branches of a single tree. I saw such a favourite tree on Fairlands estate, although at the time of my visit it did not contain more than about 30-40 combs, and these all unoccupied, showing that migration is also the rule in this locality. From inquiries it appears that at the present time there are fewer of these “ Bee tress” than there used to be, and that the number of the combs in each tree has aiso decreased. It is possible that this is directly due to the development of the coffee estates which do not favour the growth or preservation of trees suitable for these bees to build on. The combs are occasionally robbed, the trees being climbed at night and the combs cut with a knife under the protection of 4. “1k torch and a kumbli. There appears, however, to be no organised system of destruction of these combs and, where these are situated within estate limits, it is within the power of the planters to put a stop to such practices. There is a large colony of Apis dorsata underneath the arches of the Sidapur bridge and, when these combs - cea the bees are said to make the road practically impassable or traffic. “The Indian bee (Apis indica) builds in suitable hollow trees scattered throughout the wooded areas. The nests are robbed when found, but are not hunted for systematically. “ The small bee (Apis florea) builds in bushes, etc., almost every- where. The nests are robbed when found, the honey being usually squeezed out and eaten onthe spot. Itwould be impossible to protect this bee, but planters might do something to protect the nests of Apis indica on their own estates, as the taking of a nest usually leaves tell-tale evidence of chopped trees, etc.” 6. After this preliminary inspection of the areas in question, it was possible to draw up a regular plan of the experiments proposed to be done in April 1918, and the following letter was therefore sent to the Planting Expert on 24th Jannary 1913 :— (i) Districts in which experiments are to be made. “The planting communities of the Shevaroy Hills, Mysore (Bababudins), and South Coorg are the only ones which have so far signified their desire for these experiments to be undertaken in their respective areas. For the present year, therefore, it is proposed to confine the experiments to the districts named above. “JT would point out, however, that the value and interest of any experiments and observations of this nature would be considerably enhanced if they could be checked by comparison with similar experiments and observations in other districts (such as the Anamalai Hills and Travancore) where the conditions are apparently more favourable for the successful pollination of coffee. The planting communities ot these more favoured districts have not, it is true, seen any necessity for any investigation into this subject, because they are already satisfied with their outturn, but such investigations would perhaps prove of considerable value in considering the possibility of increasing the outturn of the less favoured areas. (ii) Staff required to carry out the experiments. “7 think that, to minimise risk of errors, the experiments should be carried out in duplicate in each district. We shall therefore require two observers each in the Shevaroys, Bababudins, and South Coorg. * As you are aware, I am at present under orders tu proceed to Pusa to take over the duties of Imperial Entomologist as soon as I am relieved here at Coimbatore. As to who my saccessor will be, or when he will arrive, I amas yet without any information but it is probable that he will reach Coimbatore some time in March. If this isso, it will rather upset the experiments as, if I have got to go to 2-a 12 Pusa just before the coffee-flowering season, it will be practically impossible for me to leave again at once to come down for these experiments. I might be able to take over charge by telegraph and remain down until after the experiments, but this course would require special Government sanction. In any case, at present, I can- not make any definite arrangements about my own share until I hear | reliable news about my successor. I would propose, however, that— (a) The Government Entomologist (either myself, if still here, or my successor, if he has then arrived) should carry out one set of the experiments in the Shevaroys. (b) The Assistant in Mysore (Mr. Frattini) should carry out one set in the Bababudins. (c) The Assistant in Coorg (Mr. Jonas) should carry out one set in South Coorg. ‘“* A second experimenter in each district will therefore be required to carry out the duplicate set of experiments, and I shall be glad if you will be good enough to ask the secretaries of the planters’ associations concerned to secure a reliable volunteer in each district and to let me know his name and address so that the necessary apparatus may be sent to him direct. “Should I be relieved before the flowering season and allowed to stay down here to carry out these experiments, I would suggest that I should myself revisit the Pollibetta district in Coorg and make an extra set of experiments and observations in that district. As, how- ever, it is at present doubtful whether this will be possible I cannot offer to replace one of the regular experimenters and my functions would only be supplementary to theirs ” (iii) Time of flowering-season. “I understand that the flowering season of coffee, which is very short—only of three or four days’ duration at the utmost— follows about ten days after the early rains, commonly called blossom showers, and that these usually arrive any time between about 20th March and 10th April. “ T do not know whether there is much variation in the coincidence of these blossom showers in the three districts concerned , but I would. recommend that the secretaries of the planters’ associations concerned should inform me by telegraph (a) when the banking of clouds seems to herald the approach of these showers, and (6) when the showers actually fall, giving in the second case an estimate of t which the blossom will be open. stimate of the date on “ From here the Shevaroys are distant roughl ghly one day, the Baba- budins three days, and South Coorg four days, and the Serene ins and observers should be ready two or three days beforehand. At the same time the secretaries should similarly warn the local assistants and volunteer helpers to be in readiness. “ The first information required above under heading (a) would enable me to get touring apparatus, etc., prepared and, if necessary, to recall to head-quarters any of my staff ‘ and be required for the work. 7 y staff who might then be away 13 (iv) The experiments proposed to be done. ‘The experiments proposed to be done are shown in a tabular form in an appendix. “The bags referred to will be made of two kinds of material (A) fine muslin, and (B) mosquito netting, but all will be of the same size, four feet by two feet, rectangular in shape. “ These bags are intended to be tied around selected branches of coffee-bushes for the various periods shown under the different experi- ments, to prevent pollination by particular insects at the different times quoted. [The letters (A) and (B) refer to experiments with muslin and mosquito-net bags, respectively. ] “ The experiments intended to be tried are— I. The bag to be applied before the buds open and kept applied during the whole flowering season, thus acting as a check on pollina- tion by (A) all insects and wind, and (B) all except minute insects. Il. The bag to be applied daily from dawn to dusk throughout the flowering season, the bag being removed and the flowers left exposed from dusk to dawn. This will act as a check on pollination by all day-flying insects. (A) and (B). III. The bag to be applied daily from dusk to dawn throughout the flowering season, the bag being removed and the flowers left exposed from dawn to dusk, thus checking pollination by crepuscular and nocturnal insects. (A) and (B). This experiment is the opposite of IT. IV. The bag to be applied daily from 7 a.m. to 5 p.m. throughout the flowering season, the bag being removed and the flowers left exposed from 5 p.m. to 7 a.M., thus checking pollination by honey-bees chiefly. (A) and (B). V. The bag to be applied daily from 8 a.m. to 4 p.m. throughout the flowering season, the bag being removed and the flowers left exposed from 4 P,m. to 8 a.m., thus checking pollination by all insects flying in the hot sun (e.g., mostly butterflies, syrphids, etc.). (A) iand. (B). VI. The bag to be applied daily from 4 Pu. to 10 P.M. ‘throughout the flowering season, the bag being removed and the ‘flowers exposed from 10 p.m. to 4 4.M., thus checking pollination by ‘crepuscular insects (chiefly, sphingids). (A) and (B). VII. The bag to be applied daily from 8 p.m. to dawn throughout Ithe flowering season, the bag being removed from dawn to 8 r.m., thus checking pollination -by all nocturnal insects (A) and (B). Note.--1t would have been better to have applied these bags from 8 p.m. to 4 A.M., tout it would probably be difficult to get this done at 4.a.m. As a matter of fact, there seem to be very few insects about in the hills between 4.4.mM. and dawn. i VIII. Bag to be applied during the whole time that the flowers ‘are out. The flowers of the enclosed branch to be pollinated by hand twith pollen from flowers on this same branch. 14 1X. Bag to be applied during the whole time that the flowers are out. The flowers of the enclosed branch to be pollinated by hand with pollen from a different bush. Note.—In experiment VIII the flowers will probably be best pollinated by taking and applying the pollen ona camel-hair brush, which should be carefully sterilized be- forehand by immersion in spirit. In experiment IX the flowers may be pollinated with a brush (as in VIII) or by shaking pollen from a branch takeu from another bush. In the latter case, care must be taken that no insects are shaken on to the branch which is to be bagged. In expériments V1II and IX only muslin bags will be used ag the idea is to exclude pollen carried by wind. For this reason care should be taken as far as possible to pollinate by hand when the air is calm; the early morning would probably be the most suitable time if pollen is available then. Care will have to be taken as far as possiblethat all branches bagged are free from insects at the time the bags are applied. Care must also be taken that the bags themselves are empty of insects when applied to the branches. “‘T have no information regarding any special time or stage at which the pollen grains are liberated. Possibly some of the planting community may be able to tell us about this ; otherwise it will require observation to find out the proper time to pollinate by hand in experi- ments VIII and IX. Some plants only liberate their pollen-grains within very circumscribed limits of time ; e.g., between 4 and 7 a.M. in the case of Hibiscus. The times may also vary according to local conditions.” (v) Controis, For a proper estimation of the results of the experiments, it is essential to have controls, i.e, branches exposed to exactly similar conditions except for the bagging. For each experiment, therefore, two similar bushes should be chosen beforehand, each having branches similar to one another as regards size, situation (exposure to sun and wind) and promise of flower ; one of each of these branches will be bag- ged for the experiment, the other left untouched and treated as the control of the first. (vi) Labelling For accurate identification, at. the time and afterwards, of the branches, both of those under experiment and of those used as controls, it is essential that they be plainly labelled. It must be remembered that the actual results of the experiments cannot be ascertained until some months afterwards when the berries are full-grown. For labels, I would recommend stips of smooth wood, fastened by wire on to the branches to which they relate. Ifthe details are boldly written on both sides of the labels in pencil, they should remain legible through- out the monsoon. Wood slips about 4 inches by 2 inches, planed smooth on both sides, would be of suitable size. Thirty-two of these will be required for each set of experiments ; 64 for each district. 15 As these can easily be made locally, I would suggest that the sec- retaries of the planters’ associations concerned should have them ready beforehand (by 15th March), with the necessary wire for affixing. The same number of bushes will also have to be selected in readiness beforehand and I would suggest that the secretaries or experimenters should have this done also by the same date. (vii) Observations of insects visiting coffee-flowers. “The experiments may usefully be supplemented by direct ob- servation of the insects actually visiting the coffee bushes whilst these are in bloom. ee I propose, that in each district, a representative collection of these tsects shall be made and as, for this to be done properly, continuous observation and collection from early morning until after dusk will be necessary, this work will take up one man’s whole time. For this therefore, I propose to send a member of my staff to each district ; but if this is done, arrangements will have to be made for their »ccommo- dation. They would of course provide for their own food. The men, I have available, are two assistants and two fieldmen. The assistants are Brahmins and University Graduates; the fieldmen are both Native Christians. I shall be glad to know whether districts have any particular preference for either class of men. Should I myself be free to visit Coorg (as proposed above) I shall propose to give my time chiefly to observation work of this kind, as a supplementary worker.” Summary of experiments. - - Series A| Series B eae Bags Benet ee (Muslin | (Mosquito- Object of bagging. number. coffee-bushes. bags). | net bage). I | Whole time that flower is 1 1 Check for pollination by all out. | insects. ; II | Dawn to dusk .. sis 1 1 Check for pollination by all day-flying insects. TIL | Dusktodawn .. aX 1 1 Check for pollination by ore- puscular and nocturnal in- sects. IV |7 a.m. to5 p.m... we 1 1 Check for pollination by bees chiefly. V |8a.m.to4 p.m. .. sie 1 1 Check for pollination by all insects flying in hot sun (¢9., ai butterflies, syrphids, ete.). VI |4e.™m.to8P.m... es 1 1 Chees for pollination by all crepuscular insects. VII {8pe.m.todawn .. ie 1 1 Check for pollination by all nocturnal insects. VIIL | Whole time that flower is 1 ac Self-pollinated by hand with / out.. poileon from some bush or branch that is enclosed in this one bag. LX | Whole time that flower is 1 1 Cross-pollinated by hand with out. pollen from a different bush. Series A is especially to prevent pollination by all insects and wind. Series B is especially to prevent pollination by all large insects (¢.g. , honey-bees). 7. In accordance with my proposals, it was arranged that a duplicate set of experiments should be done in each district, and that a member of my staff should also be sent to each district to observe the insects frequenting the flowers. The following were the ubservers :— Experiments. District. Observing insects. Original. Duplicate. Shevaroys ..| Government Entomologist, | C. Dickins, | Y. Ramachandra Rac, Yercaud. Nagalur. Entomological A ssist- ant. Bababudins ..|G. Frattini, Assistant | N. G. B. Kirwan.| P. Susainathan, Ento- Scientific Officer. mological Fieldman. Coorg (Sidapur). | W. Jonas, Assistant | G. Newbery T. V. Ramakrishna Scientific Officer. Ayyar, Entomo- logical Assistant. 8. In spite of the fact that these experiments were arranged for in duplicate, unfortunately it happened that results are only available from two of them—those carried out by myself and by Mr. Kirwan. Mr. Dickins’ experiments were done carefully but a subsequent very heavy hail-storm did so much damage to the bushes that the experi- ments were nullified. I understand that Mr. Newbery did carry out experiments but he has not sent in any results ; from a brief conver- sation with him during my recent visit to Coorg, however, I gathered that his results were on much the same lines as those done by myself and Mr. Kirwan. Neither Mr. Frattini nor Mr. Jonas carried out the experiments, su far as 1 am aware—at least, no results have ever been forwarded. 9. The Yercaud experiments—Having received advice that the coffee was about to blossom, I reached Yercaud on 23rd April 1918 and found the flowers still in bud and not yet opened. Suitable bushes were selected and branches were carefully covered with muslin and mosquito net bags, as previously arranged, and the branches labelled with permanent wooden labels ; other branches, in each case on the same bush and as similar as possible in size, shape, position and promise of blossom, were also selected as controls and labelled. The earliest flowers coramenced to open on 25th April and the regular routine of applying and removing bags was started and continued regularly until 8rd May, by which date all the flowers wore over and withered, and the bags were then removed from all experiments except Nos. IB, VIII and IX which were unbagged on the next day. At this time all the flowers were quite over, all withered and many dropped off; no bees were to be seen on the bushes andit seemed ae whether the few odd late flowers were still yielding any 10. I revisited Yercaud on 17th Decemb 1 er 1913 % results and collect the berries. These, however, da grassy uniform in ripeness, some being rive and dropping off the bushes others still quite green and unripe, and afew not yet full-grown. To 17 obtain strictly uniform results, all the berries should have been picked when properly ripe. ‘lo have done this would have entailed my re- maining at Yercaud tor at least three or four weeks, visiting the bushes every day to pluck the berries as they ripened. I did not feel justitied in doing this, more especially because plucking had already commenced and there was considerable danger that an ignorant cooly, by stripping the bushes under experiment, might have spoi't the whole series. ‘l'here was also danger of losing the ripening berries owing to the attucks uf birds and other auimals and: also to their dropping off under the influence of the cold wet days which were experienced at this time, I therefore picked all the berries of the experimental branches and of their controls, noting their comparative appearances as under :— Resutts of experiments at Yercaud. Number of experi. ment. Bag applied to flowers. Appearance of berries when plucked. Remarks, I-A -B Se en Hea ..¢ Nom na III-A ,,< All the time .. Contro) All the time ., Contro} Dawn to dusk, Contro] ve Dawn to dusk. Control Dusk to dawn. Control 28 plump berries, quite rimi- lar in appearance to norma! (un-netted) ones. All green except one nearcst tip of branch, which is turning red. 24 berries, all plump, of which 7 are turning red. 22 plump berries, only one turning red. 23 bersivs, of which 3 are less | than normal size and 2 are reddened. 68 berri-s, mostly plump and normal and apparently equal! to those on untre- ated bushes, but 8 are much smaller; 22 red- dened. But 40. more berries appear to have dropped or been + lucked recently 8 they are only represented hy broken stalks, so thet the real number of berries deve- doped was approximately equal to control. 107 berries, of which 14 are smallerthan normal and about 28 reddened. 87 berries, of which 4 are sma.J-sized and 7 red- dened. e 83 berries, of which 3 are small-sized and 6 red- dened. 65 berries, mcstly green, of which about 20 look very smull and only 4 rediered. ‘She whole branch locks very poor. 72 berries, of which about 12 are small and only 3 yeddened. The whole branch looks poor. i 9 ome & There is apparently nothing !o ch..ose bet- ween experim: nt» I-A and 1-B and their euntrols. No apparent difference between JI-A and its control,- allowing in the former for bro! en- off berries which ure probably due to birds, Apparently little differ- ence between experi- ment 11-B sand oontrol. Little difference between TiI-A and control is perceptible. 18 Appearance of berries when plucked. Remarks. : ied t af oxperc | Beganpli to ment, (| Dusk to dawn. III-B ..¢ Control 7 a.m. to 5 p.m. IV-aA .. Contro) 33 IV-B .. Control ne V-A V-B... (| 4 a.m. to 8 p.m. | Control VI-A ..< ¢ 4 P.M. to 8 a.m. vis Control one L (| 8 P.w. to dawn. VIL-A..¢ Control as . VII-B VIEL-A ..| Whole time (self-po)li- nated), IX-A ..] Whole _ time (crosspolli- nated), | 149 berries, p)ump and healtby- |} looking, of which about 12 are small and only 2 reddened. 87 berries, of which about 9 “are smaller and only 2 reddened. 38 berries, all large plump and 22 reddened. 50 berries, of which 6 smal) and 6 reddened. and Nil Nil as or ee Only one small green unheal- thy-locking berry. Label had been moved, but the wire which had fastened it was found on a braneh (of same bush) with 3 unheaithy- looking berries. 104 berries, fairly plump on the whole, about 7 small, 1 rotten, 4 reddened. A healthy-looking bran- h ch, A rather small branch with 3 berries, of which 3 small, rest plump and 20 reddened. 56 berries, rather below uve- rage size, 8 decidedly small and 8 reddened. 38 berries, rather below ave- rage size, 11 small, 7 reddened. Nil ee oe 68 berries, of fair average size and plumpness as com- pared with others on the same bush; of these about 7 are smal] and 14 reddened. 92 berries, all green and plump and of fair ave- rage size as compared with otherson the sume bush ; of the 92, 2 are slightly smaller and 2 reddened. | WN id Both III-B and its eon- trol are healthy-luok- ing branches with little to choose bet- ween tl em. IV-A looks better and healthier than its contro). ‘ The labels had been removed and therefore the tranches could not be identified. These experiments, originally designed to test the influence of inseots flying in the hot sunshine between 8am. and 4 P.M., were abundoned aa impracticable. « Abad bush with very few berries anywhere. Experiment 4 total failure. Little to choose’ on the whole bet ween general appearances of VJ-B and its control, bat the latter was riter— perbaps it was nearer an open space and slightiy more exposed to the sun, Little to chooge between VII-A and control on Senera] uppe -rance. Only the contro] branch could be found. The label of ViI-B had been removed and the branch could not be identified. 19 ies ee . nad wT Number Bag applied to Appearance of berries sori- neta elo Ta flowers. when plucked. : Berets a“ X-B.. | Whole time .. | 11 berries, mostly plump and | A dapitoute ‘o tL By ‘a : green and of fair ave-| mosquito-net bag: being rage appearance incom-| applied all thetime. the parison with others on| flowers were ort. oad the same bush; of.the ae 11, 3 are reddened and |. ‘ et ic of these 2 look poor. hes 11. The berries, both of the branches under experiment, and! of, their controls, were sent to Mr. Anstead, the Planting Expert; who in: his letter No. 94, dated 29th January 1914, wrote to me. ‘regarding them— 8 od qeieaia ‘T have to inform you that I have finished the eartn of ' the’ coffee berries. you sert me from the Shevaroys. The berries: were picked far too green to cure well and | think that nothing canbe’ judged from them as to quality. eel “‘T send you the resulting berries under separate cover aindh you’ will see that the result is poor. When all are so bad it is impossible to grade them at all, beyond saying that I-A control, II-B and: ives are in best and perhaps III-B and VI-B the worst.” 0 +o With regard to Mr. Anstead’s comment that the, ‘berries’ ** were D aearal far too green to cure well,”’ I quite agree that this was so, but this seemed to be the best thing to do. To have remained until all were ripe, it would have been necessary to have prolonged my’ stay in Yercaud at least another three weeks, visiting the bushes daily’ or even more often and risking loss of many of the berries owing. to’ their dropping prematurely or being detached by pickers or birds or other animals. If I had bags with me, I might have reapplied them to the branches to catch the berries as they dropped, but the’ ‘chances: would have been very large that the bags would have’ ‘disapr eared: in’ toto with the pickers working in the coffee all around. I therefore’ considered that the best thing to do was to pluck the berries as they were, to get comparative results regarding the number and size of the berries. 18. The Bababudin experiments. —These were carried out by ‘Mr. N. G. B. Kirwan and are detailed in his letter of 16th J ea AVES 0 the Planting Expert. A copy of this is given below :— eh ile pee ac “ ARaptpacooL Hsta1z, _ Chikmaglur P.O., 16th Pray 1914. “ R. D, Ansteap, Esq., oe ore vy Planting Expert, Bangalore. “ Dear Sir, be ats Bees and coffee fertilization. “Under separate cover I send the samples of coffee obtained from the above experiment. I regret to say No. 5-B is missing, ¢ as on pick- ing the crop [ found this label was missing. 4 ‘‘ The blossom showers fell on April 18, 20 and 21, totalling 1:10 Inches. “The bags and labels were applied on the evening of the 2Ist. The experiment commenced on 27th April 1918 and concluded on 29th April 1913 when the bags were replaced fora week. During the course of the experiment the following points were noticed :— ‘A very large number of bees were about owing to large hives of Apis dorsata on the cliffs above the estate. Fifteen to thirty bees were constantly counted ona single tree by the Government Ento- mologist’s fieldman and myself, the proportions of the different kinds of bees being about 55 per cent. Apis dorsata, 35 per cent. Apis indica, 10 per cent. Apis florea, “The bees completely nullified the experiments to check fertiliza- tion by other insects as they entirely failed to keep to their scheduled _ time. For instance, one bee made a determined effort to attack the branch I was hand-pollinating at 6 a.m. (experiment 8), For experi- ment 5, the bag should be removed at 4 p.m. At 5 p.m.on the 27th there were five bees on the experiment tree, viz., three Apis dorsata, two Apis indica. ‘This bag had been removed at 4 P.M. .“In general I especially noticed bees about from shortly after dawn until it was too dark to distinguish them any longer. From dawn till 7 a.m. and 4-80 p.m. till dark the majority of the bees were not out. “The number of other insects compared with bees was negligible, A point noticed while cross-pollinating experiment 9 was that flowers from a branch of another tree that had been open to insects the day before fell when shaken. The experiment branch, which had been enclosed in the bag the day before as some of the flowers were not out, did not fall when shaken. “The two trees selected were healthy young Coorgs, about 10-12 years old. As they were carrying a heavy crop a lot was pruned off. {n spite of this 1 regret to say they developed leaf disease in December, and thus to a great extent prevented comparison of the different samples. ‘As will be seen in the tabular statement below the quality follows the leaf disease. The branches picked on 7th January were not much affected by the leaf disease. The rest were. “IT had originally intended to count the berries after the blossom, so as to check the loss from the commencement, as should undoubtedly have been done. While carrying out the experiment I found it impossible to entirely eliminate damage to the flowers by the bags. As it was impossible to say accurately what flowers had been knocked by the bags, I did not do this. The accompanying table records the following points :— A. Muslin bag B. Mosquito- net bag. C. Control. Column 1. Number and class of experiment. ie 2. Time bags were applied. 3, Date picked. J. 7 Ge Jans 39 anuar or Ja 9 : ” 4, Percentage of ripe frint nuary 26, 1914 Column O WN Sor 21 . Number of berries after drying. . Number of beans of clean coffee. . Number of pea berry. . Actual total of beans. . Possible total of beans (As one normal berry contains two beans but apea berry only one, this a to be allowed for in counting the loss.) oss. » . Percentaze of loss. . State of branch at time of picking. . State of sample coffee. 22 91-G | set! oso’s | 6te's | ¥2e 969°2 | ZoL‘L ' ‘i “Treaig | °° ‘od r6o-8 | 8 OL el rd 104 68 ss 119% ‘od 6) : “(pezeurrjod “Aap pus treo £104 | °° peop pug | 48-2 |e | got Ol of g0l +9 #2 Wee | 88029) emg efeq AN |"V ‘6 ‘od ‘op “oT 89-1 |e 8Lt G11 9 691 26 us 49% ‘od fe) dn partrp qouerq i *(poqeutrfod “POTTSAltgs pus [leurs | yseT ‘peop pug| sre | ¢ 6ST FIL I ral 8 8st ms | #98) euNy ofomm | “8 ‘od : “7 0 IL’ | 681 81 g 81 96 ag a4 7 ‘od is ‘og "0g L6-8 | 9Lt 691 (a4 #1 66 19 qdZ a “oa ‘a “Teno N | ** “* TSWION | 93-9 | L rau GOL we yA oh 99 aye “* UsEep—Ard BV TL “petteatigs pue rems Aroq | * apeospyrog | ys.z | g $L) 69T $1 oor $6 rd ayo | 0g fe) “poj[eaizys pus |[sur~ | *~ peep pu | 16 T 60T 80 = LoL eg ayn | 7" ‘od ‘a “TRUION | °° og 0r-g | 9 ITI gor Il 6 19 oF m92 | ‘Wd gB— Wd FV 9 “porleatiys ouos f [Twug |} ** esvesrpy ey | +6.2 | $ get sl 9 921 x Si waz | °° ‘od ue) eae 3 Set At ae . . . oe ae . ee U44G —- ‘og ‘g “Por[eaqsys outos feztspoop |-* esvesrpyeoT | 99.9 | 6 | get 921 1% 66 18 1g a9 | "0 “Wa FHV BV G ‘od ; ‘od Lut | * 99 6g 8 Lt 78 8! TL ve ‘od ‘0 “og ‘og eL-€¢ | 99 | 99T Sol 8¢ 02 01 19 Lee) 2 og “@ : “eadoN | °° [eUdL0 N : “| te 1G 8 oF og” 98 TL wa g— wy, | YF “pa[eatiys pus [[eo's A190 4 Peep pum | Glee | g¢ €81 82! ee £6 38 g 493 | 7° od me) “VeunI0 NT Teuton | 99-2 | + Out OFT (ad Il 98 cL WL pal ‘OT ‘a “Peleattys puy [peur | - prop pu} 29-T | #8 Isl 0% 191 zor |st- yyoe | umep-4enq | “V “e “paTiaatas ; ‘OT ee A OL1 ZG OFT 86 Lg 1395 | 7 ‘od 0 ea B yq ezIB poop eeBestp JOT | 99.2 |g | GGL PAT 14 Li oll Lt qyae | 7 °d a ° : peep pum | see | F Tal LIL @ FIt 29 ce q9e | * yeup-umed |"y & ‘od Oi] ion eke aie) 89 1 29 78 1g UsL as ‘od ue) ‘OT oq 8h-3 0 | 1 1F 0% 1 68 1z vol WL = ‘OT ‘d STS SSON | (POTION | G6-4% | 131 18 09 L &G $F 98 WL ae emt} sou |W ‘T ‘B0V ae I Il 1 6 8 L 9 ¢ + & z 1 “ABO JO F R *Aszeq | ‘suveq | ‘sorzaeq | *ymmsz yo! -eidureg *qousta Jo 93849 gavnres BsorT] ie coat ved Jo qo jo es e\ued es sporidds deg _ “19g Taised | TD | aqumy | soqainy| requmy| -108y a 23 *"I'here was an undoubted loss of crop on the two branches that were not fertilized at all, 2.¢., berries that never swelled at all. Yours faithfully, N. G. B. Kirwan.” 14, So far as it concerns the actual pollination of the coffee plant, the foregoing experiments appear to show (1) that the aid of insects is not essential but that fertilization can and does take place when all insects are excluded during the whole period that the flower is out, and (2) that wind, and similar natural agencies, are also not essential factors, since the muslin bags used in experiment I-A were of sufficiently fine material to exclude pollen altogether. These results were certainly very unexpected, but some light on the subject seems to be afforded by Dr. von Faber’s experiments of which I only became aware after mine had been done, when Dr. Coleman published the following note in the “ Planters’ Chronicle” (Volume VIII, pages 350—351; 19th July 1913) :— “In 1910, my friend Dr. F.C. von Faber published in Dutch some notes on the biology of the coffee flower in which he gives the results of experiments performed by him on fertilization. He experi- mented with Liberia coffee and under Javanese conditions so that these results cannot be taken directly as applying to South India. They however suggest a line of investigation which might, perhaps, be profitably followed here in studying the question of the importance of cross-fertilization. “ Dr. von Faber first examined the question of pollination and found tbat out of 78 flowers of Liberia coffee examined, in all but eight self-pollination had occurred in the bud before the flowers opened ; moreover the pollen grains had in many cases, already germinated at the time of opening of the flower and the pollen tubes had begun to grow down through tissues of the style towards the ovaries. * From this result one might be tempted to conclude that, in the case of Liberia coffee, cross-fertilization can occur in so few cases as to. be quite negligible. This is, however, not the case, as, on further investigation, Dr. von Faber found that, where pollen from other flowers on the same plant or from uther plants found its way to the stigma and germinated, it grew about twice as fast as the pollen from the same flower. There thus occurs a sort of race between the foreign and the “self” pollen tubes as to which shall fertilize the ovaries and the results depend entirely upon how early the foreign pollen reaches the stigma. “Dr. von Faber has also some remarks to make on the réle of insects in the fertilization of Liberia coffee flowers. He states that, although the number of insects in a coffee plantation is usually small in comparison with the number of trees, this is to a certain extent equalized by the rapidity of insect movement, whereby a large number of flowers are visited in a very short time. Bees seem to have been among the most prominent insects observed by him. The following observations made by him are of interest :— 6th April—Tree No. 8—At 8 a.m., 48 flowers found open; within 15 minutes, 32 of these were visited by insects. 25th April.—Tree No. 7-a—At8 a.m. 72 flowers found open; within 25 minutes, 48 of these were visited by insects. 24 14th May.—Tree No. 15--At 8-30 a.m. 64 flowers found open ; within 45 minutes, 52 of these were visited by insects. 14th May.—Tree No. 18-b—At 10-30 a.m. 88 flowers found open; within 15 minutes, 24 of them were visited by insects. “The visiting insects were, in each case, examined and were found invariably to be covered witb pollen. ‘The paper under reference also contains an account of observa- tions on the robusta varieties of coffee, where it was found that self- fertilization in the bud rarely takes place. ‘Ihe author found here also that foreign pollen grains grow more rapidly down through the tissues of the style than do the ‘self’ grains. ‘‘ With regard to Liberia coffee, the author concludes that cross- fertilization was the original method and tbat self-fertilization has arisen later to guard against failure in fertilization. “ Ag far as I am aware, no similar studies have been made on coffea arabica either in India or elsewhere and it would, I believe, be a matter of considerable practical interest to ascertain whether a difference in the rate of growth of foreign and ‘self’ pollen grains exists here similar to that described by von Faber for Liberia coffee in Java.” 15. Dr. von Faber’s experiments show that the plant itself expresses a decided preference—if such a phrase is admissible—for foreign pollen; or, in other words, cross-fertilization is secured when- ever possible, self-fertilization being kept in reserve as a last resource. The experiments in the Shevaroys and Babadudins show that there is little difference in the quantity of flowers fertilized by insect-visitors and Dr. von Faber’s paper explains (no doubt correctly) why this is so, but this does not necessarily mean that there is no difference in quality between berries from flowers which have been cross-fertilized and from those which have been “‘selfed”. The quantity of beans collected from the experiments is too small to ensure a satisfactory comparison of quality, but the endeavour made by the plant itself to secure cross-fertilization strongly endorses the idea that such cross- fertilization is highly desirable—a view which is of course in accord- ance with experience gained in the case of other plants. We may conclude, therefore, that the visits of pollinating insects should be encouraged as far as possible. 16. Turning now to the insects which secure such cross-fertiliza- tion in coffee, before carrying out the experiments, I endeavoured to find out what information was available regarding such insect-visiters of coffee in India, but the information available proved to be most remarkably scanty. Knuth’s ‘‘ Handbook of Flower Pollination”. to which one naturally turns for information on sucha subject, gave very little information on this point, as is shown by the following extract (English translation, Volume IL, page 549) :— “ Coffea arabica, L.—According to Bernoulli (Bot. Ztg., Leipzig, XXVIII, 1869, page 17) only small purely female fertile flowers are present at the beginning of anthesis, Ernst states that the herma- phrodite flowers are protandrous. “ Visitors—Bourdillon chiefly observed Lepidoptera (Nat: London, XXXVI, 1887).” picopters. (Natare, a 25 17. Reference to Bourdillon’s original paper, however, showed that this had been considerably misquoted by Knuth, as Bourdillon referred specially to bees and mentioned Lepidoptera only as subsi- diary agents in pollination. The following is a copy of the original letter (“ Nature” XXVI, 580; 20th October 1887) :— “ T send you the following notes on the fertilization of the coffee plant (C. arabica) which I made sometime ago, and which may be interesting to those who study the subiect. “ Your readers are doubtless aware that coffee was cultivated some twelve years ago to avery large extent in Ceylon and South India, but owing to the attacks of leaf disease, the area has been rapidly reduced, except, I believe, in some parts of Coorg and Mysore, where the climate is drier, and the leaves suffer less from the fungus. It has now been largely replaced by tea. ‘The jasmine-like flowers of the coffee are borne in clusters in the axils of the leaves, and appear simultaneously all over the estates. After a prolonged drought of one or two months, or even more, at the beginning of the year, there is generally a heavy fall of rain, sometimes lasting only an hour or two, sometimes continuing for two or three days: the amount that falls must be enough to saturate the ground, and should not be less than one inch. “ In from six to eight days from the time of the first shower, the flowers burst into full blossom, last for a day, and then drop off. On the evening before the blossom is fully out, if the flowers are examined it will be found that they are partially open, the stigma being pro- truded and receptive. During the night the hum of insects can be distinctly heard, and I am of opinion that the flowers are largely fertilized by night-flying insects which carry pollen from those flowers which happen to be open rather before the others, as some are delayed. On the following morning all the flowers will be found open, and the field of coffee presents a sheet of white. These flowers are frequented by immense numbers of bees, of two kinds, one about three-quarters of an inch long and black, the other smaller and with white bands round its abdomen. The stigmas now are covered with pollen, and the anthers bursting, and the larger of these bees may be seen buzzing from flower to flower sweeping up the grains of pollen between its front legs, and rolling them, into balls. Long before evening all the anthers are exhausted of pollen, and the insects have departed. Besides bees, some butterflies visit coffee, such as Hypolim- nas bolina, Papilio polymnestis, and two or three Danaide. “The coffee plant by being proterogynous is intended by nature to be cross-fertilized but owing to all the plants in one clearing being usually grown from seed of a single estate, there must be a great deal of interbreeding, more especially as all the coffee of Ceylon and most of South India is supposed to be descended from a single plant introduced into Batavia about two centuries ago. This may have something to do with the manifest deterioration in stamina of the younger coffee. Quiton, S. TravancorE, Inpra, T. F. Bourpinton.” 18th September 1887. 26 18. The larger bee referred to above is evidently Apis dorsata, the smaller one Apis indica. 19. My own observations on the insects frequenting the coffee flowers at Yercaud in April 1913 were necessarily scanty, as my time was occupied with the experiments, but the following notes were made :— Apis dorsata (the large rock bee) was in comparatively small numbers, collecting pollen in the morning as a rule, and nectar in the afternoon. One which was watched when gathering nectar visited ten flowers (all on the same bush) in one minute.— Another specimen, which was collecting pollen, was observed to visit about 22 flowers in One minute, this number including those which it crawled over and might therefore have fertilized; in a second minute the same bee visited about twenty flowers. When collecting pollen the bees visit far more flowers (approximately twice as many) than when collecting nectar, as in the former case they often crawl all over a bunch of six or seven flowers, whereas in collecting nectar they generally fly straight from one flower to another. There seemed to be more speci- mens of Apis dorsata on the flowers towards the end of the flowering period than at its commencement. Apis indica (the Indian bee) was fairly common, appearing on the wing at sunrise (6°0 a.m.) and working until after sunset. They gathered pollen chiefly in the morning and nectar in the afternoon. When gathering nectar about ten flowers per minute are visited, and about twice this number when collecting pollen. When sucking the nectar, Apis indica often touches the stigmas with its legs as it arises from the flower. Its habit of visiting flowers: erratically (as often as not flying off one flower on to another one ou a different branch, albeit of the same bush) has doubtless considerable value in securing cross-fertilization. When sucking nectar, Apis indica does not quiver its wings, but remains motionless except for extension and retraction of the abdominal segments. When collecting pollen, the bee usually does not settle on the fnwer but hovers over it and scrapes the pollen off the stamen by embracing it with its mid-legs and drawing these upwards, thus rubbing off the pollen which sticks to the legs and is then scraped off and passed to the regular pollen baskets of the hind-legs. In pollen-gathering there is more chance of fertilizing the flower because the legs hang down when the bee is on the wing and may (and sometimes do) touch the stigma when moving from one flower to another or hovering over a flower scraping up the pollen- grains. Apis florea (the Little Bee) was not common and no exact details ues meee: but their whole bodies appeared to be covered with pollen. Xylocopa tenuiscapa (a large blue-black Carpenter Bee) was fairly common and very active in visiting the coffee flowers, although T noticed some which were busy on Potato (Creeper (Solanum) to the entire exclusion of the adjacent coffee, These large bees sprawl all over the flowers, carrying pollen all over their ventral surface : the often go right up or down a twig, crawling over every flower ee probably pollinating every blossom which they touch. 27 20. Butterflies seemed to be unimportant as pollinators. Danais genutia was seen to visit the flowers and to touch them with its legs and may pollinate, and Hypolimnas bolina and a small brown Hesperid were also noted to feed on the nectar. A MWeptis was common amongst the coffee but settled by preference on the leaves and was not seen to visit the flowers. A few other butterflies, belonging to the genera Junonia, Ypthima, Mycalesis and Terias, were also noted amongst the coffee but not as flower-visitors. 21. A day-flying clear-winged Sphingid moth, Cephomodes picus visits the blossom and feeds on the nectar with outstretched tongue whilst it hovers on the wing, but its rapidity of flight and wariness made exact observations difficult. It did not, however, appear to touch the stigma at all with its tongue. This moth, it may be added, lays its eggs on coffee, the caterpillar feeding on the leaves. 22. After dusk a few Sphingid and other moths visited the blossom. They could generally be heard rather than seen, but their number seemed very small—certainly less than I had expected. 23. A few specimens of a yellow-spotted Cetoniad beetle, Clinteria auronotata, were found on the bunches of flowers, where they were licking up the pollen systematically by the aid of the brushes on either side of the mouth. Probably these beetles thus eat up a good deal of pollen (as the bees also collect it) but do good by sprawling over the flowers; however, no pollen was visible on the bodies of the beetles. #4, The other insects noted were in such small numbers that they may be regarded merely as casual visitors of little practical importance so far as pollination is concerned. 25. There was no doubt in my mind, from what I saw of the insects which were frequenting the blossom, that the bees (Apis dorsata, apis indica, Apis florea, and Xylocopa tenuiscapa) are the only really important pullinators of coffee. The movements of the Honey-bees (Apis spp.) are generally rather erratic, as they buzz apparently at random from one flower or branch or bush to another, thus securing a large measure of cross-pollination, whilst the Carpenter-bees (Xylo- copa) are most systematic in their mode of working over the bush, whilst they are literally dusted over with pollen and sprawl about over the flowers. 26. I attach the reports on the insects noted on coffee-blossom in the Shevaroys and in Coorg by the two Entomological Assistants sent to those localities for this purpose. (1) Report on a tour to Yercaud, Shevaroys, by Y. Ramachandra Rao, Entomological Assistant (Aprii-May 1913). “ Under orders of the Government Entomologist, I started from Coimbatore on Sunday the 2Uth April 1913 by Mail for Yereaud via Salem. I reached Yercaud the next day at about 9 a.m. The object of the tour was to make accurate observations as to which insects visit coffee flowers, when they are most active and whether they are really useful in the fertilization of the flowers. When I arrived at Yercaud the branches of coffee plants were covered with plentiful 28 flower buds in various stages of growth, but none of the blossoms had yet opened. In fact, the first flowers opened only on the morning of the 25th April. By the 27th, all the estates in Yereaud were in fall bloom and presented a sight of wondrous beauty with their exuberant masses of pure white sweet-scented flowers. Flowers generally faded away on the 4th day, so that by the lst May most of the blossoms had either dropped down or faded, but scattered bushes that had burst into blossom later still had fresh flowers on the 2nd and 3rd May. «2, Observations as to flower-visiting insects were made on and off between 6 a.m. and 7 p.m, during daytime and as far as possible, speci- mens of insects noted were collected. On two nights, I examined coffee bushes between 8 p.m. and 9 p.m. with the aid of a lantern, but though I found a few moths flying about, I was unable to ascertain whether they came to flowers at all, as, disturbed by the brightness of the light, they either came to it or flew away. My observations, therefore, as regards the night-side of the question are scanty. “3. The following were the flower-visiting insects that I noted at Yercaud :— Hymenoptera :-— (i) Bees :—(1) Apis indica—the most common of the bees at Yercaud, It was found on flowers from 6 a.m. up to nearly 6-30 P.M. It was the only bee that was found at allin plenty. It was, however, not present in sufficiently large numbers, there being generally not more than 2 or 8 on the average on each bush at a time. This bee was at times found confining its attention to nectar and at other times merely gathering pollen. (2) Apis dorsata.—(The Rock Kee). This was rather scarce at Yercaud. on coffee flowers. I did not notice it at all in some estates and where present it was not plentiful. It was most active at midday. (8) Apis florea : found in small numbers on most estates, but is not so scarce as Apis dorsata. Like Apis mdica, this bee was noted confining its attention at times to nectar alone and at other times busying itself solely with pollen-gathering, every anther being regularly scraped up for the purpose, (4) Halictus sp.—Noted only on one occasion. It was collected while gathering pollen on a coffee flower. (5) Ceratina sp.—Only two instances of this bee visiting coffee flowers were noted. It was common, however, on certain small labiate plants in open places. (6) Anthophora sp. found visiting coffee on only one or two occasions. Specimens could not be secured. (7) Xylocopa tenuiscapa.—'\he Large Black Bee or the Carpenter Bee: fairly common—though not, by any means, even ds numerous as Apis dorsata at Yercaud. Being a big insect and of active habits, it visits a larger number of flowers at a time than any of the Honey-bees. It is, however, not a sysiematic worker; forafter visiting flowers on a few branches of a bush it rushes off to another situated at a distance whereas the fl oney-bees (especially Apis indica) do not, as arule, leave any flowers unvisited before departing for the next bush. : (11) Wasps.—I observed two instances where (1) a Sceliphron sp. and (2) another Sphegid visited coffee flowers and as far as I could observe, went in for the nectar, but 1 was unable to catch them. 29 (ili) Ants.--I noted, at least, 6 species of ants visiting coffee flowers. Of them (1) Myrmicaria brunnea and (2) Camponotus paria (workers minor)—both rather large ants with long legs—were found walking over the flower clusters and occasionally stooping down to the mouth of the Corolla—though the large size of their heads could not have permitted of their reaching the nectar in the nectaries. It is just possible that their long legs might be instrumental in the transference of the pollen from the anthers to the stigmas. The other ants were :— (3) Tapinoma-melanocephalum.—This ant was present in large numbers on some bushes and visited the corolla of flowers. Being a tiny ant, it obtained access to the nectaries, but as it did not generally climb up either the stamens or the styles, it seems to have no significance in the fertilization of coffee. (4) Monomonium gracillimum var mayri was present in some numbers on a few bushes. ‘l'hey came up froma hole in the ground up the stem and branches and entered the corolla of flowers in search of nectar; they did not run up the stamens or the styles. (5) Sima attkeni (?)—On some bushes I found a few of these slender elongate ants running up and down leaves and flowers, but though possible it is very doubtful whether they actually transfer pollen to stigmas. (6) Acanthoiepis frauenfeldi—Only two specimens of this ant were obtained, one inside the corolla and the other on the petals of the flower. (7) Lepidoptera——Four different butterflies were noted by me settling on flowers and sucking nectar up with their proboscides, but so far asI could observe they do not seem capable of carrying pollen from one flower to another. Moreover they seem to visit coffee flowers but rarely. (1) Hypolimnas bolina —noted on three separate occasions. (2) Mepiis ewrynome—on two occasions (°) Danais nilgiriensis—on three occasions. (4+) A Lycenid (which I was not able to catch). (iv) Mcths.—(1) Cephonodes picus.—Observed to be present in most estates. Itis a dayflying moth and was to be found at all times of the day from 7 a.m. to dusk. Itis very active and flies about with a characteristic humming sound. It sucks up nectar while on the wing and flies from flower to flower very quickly. While visiting flowers, however, the tongue—while it may touch the anthers—does not reach the stigma and it is therefore doubtful how far it is concerned in the fertilization of coffee. When buds are just about to open, however, the stigma alone projects slightly out from the midst of the petals and at this stage, I observed Cephonodes thrusting its tongue in (in buds not yet open, it actually forces the proboscis in) and in this case the moth would seem to act as a carrier of pollen. (v) Along with the blossoms, fresh shoots are also put forth by the bushes after the first rains. J observed the Cephonodes -moths depositing smali yellowish green spherical eggs on the tender buds or young leaves of coffee. Some eggs collected by me hatched on the fourth day into greenish Sphingid larvae with a conspicuous strai-ht black horn They were fed on tender coffee leaves in breeding cages, but they did not somehow thrive on them and died, 30 (vi) Colleoptera.—(1) A Oetonid beetle—Clinteria awronotata— was collected by me froma bunch of partially faded coffee flowers. Probably it feeds on the anthers and stigmas: but I do not think it has much imporiance as regards the fertilization of coffee flowers. ° (2) A Nitidulid of small size was collected from inside the corolla of a coffee flower. ‘ (vii) Diptera Syrphide—three different kinds were noted on coffee flowers :— (1) Syrphus sp. (?} very much like the Syrphus found on Aphis in the plains. This was noted on three different occasions on coffee flowers, where the pollen was licked or scraped up with the tip of its tongue from the anthers and later on the style was similarly treated. (2) (?) Eristalis sp. A large fly resembling a bee to a certain extent. It was only once observed by me on coffee flowers—on which occasion I noticed it brushing pollen up with the tip of the tongue, visiting all the anthers by turn. It also licked up the surface of the stigma and later thrust the proboscis down the corollar tube for the nectar. Certainly a useful insect if present in sufficient numbers. (8) Another Syrphid found also on 4 flower on one occasion. (4) Muscidae : A fairly large fly—once noted on aflower— thrusting the proboscis down the corolla--apparently after nectar. (viii) Hemihtera— Pyrrhocoridae (1) Serinetha abdominalis.—This bright orange-yellow bug was notedin large numbers in some coffee estates. Usually they were found either on coffee bashes or on fallen leaves on the soil. Some were noted on coffee flowers as well, but, as far as I could observe, they made no attempt either to seek the nectar or to suck out the sap of flowers. , (ix) Aphide, (2) Aphis—About five specimens of Aphis were found on a single unopen flowerbud. They were tended by a species of Cremastogaster. “4. It is quite possible that, in addition to the insects enumerated above, there might have beena number of insects with decided flower- visiting habits that might have escaped my observation. Especially is this likely to be the case with moths and other nocturnal insects at which time my observations are confessedly scanty. Nevertheless, I feel pretty certain that coffee flowers (setting aside the probability of wind-fertilization) are in the main fertilized by the action of the day-flying visitants of flowers. Of the list of about 29 insects enumer- ated by me in this report, the really important ones are certainly the following :-— a The three honey-bees— Apis indica, Apis dorsata and Apis ored. (2) Xylocopa tenuiscapa and (3) Cephonodes picus. _ “5, Of the three honey-bees, it was only 4 pis indica that was found fairly abundant. Apis florea was present in but scanty numbers while Apis dorsata was absent altogether in some estates I examined. In the cotfee flower the pistil is very elongate and the stigma stands well separated from the anthers, so that when Apis indica and Apis florea visit the flower they are unable—whether while seeking nectar or 31 while gathering pollen—to reach the stigma. As, however, coffeo flowers occur in thick clusters pollen seems to be deposited on the stigmas as the bees step out from one flower to the next. Apis dorsata being a bigger insect seems to be better adapted for the fertilization of the coffee flowers. Xylocopa is still more efficient in this respect— as by reason of its weight the style is bent down so that the stigma comes in contact with the bushy posterior tarsi which carry abundant pollen. “6. Cephonodes—-so far as its visits to fully open flowers are concerned— seems to be of little use in fertilizing flowers. Some pollen may certainly get attached to its tongue while sucking nectar, but the stigma is situated so far above the reach of the proboscis that fertilization through that means appears to me quite impossible. The moth has the habit, however, of visiting just opening flowers where the stigma, which just protrudes out from the flower, is certain to have a few pollen grains deposited on to it as the tongue is thrust in for the nectar. All the other insects appear to me of little significance as compared with the honey-bees, the Carpenter-bees and the Hawk- moth. Y. Ramacuanpra Rao, 23rd May 1918. Second Assistant.” (2) Report on a tour to Coorg, in April 1913, by T. V. Ramakrishna Ayyar, Entomological Assistant. ‘© On the 17th April a telegram was received from Mr. Jonas, the Scientific Assistant to the Coorg Planters’ Association, regarding the flowering of coffee. Under orders from the Kntomologist, I left for Sidapur the next day and reached the same on the 22nd morning having been unavoidably delayed at Mysore for two days for want of conveyance to Coorg. “ 2, On reaching Sidapur, I heard from Mr. Newbery of Margalli Estate, who was conducting one of the experiments in connection with ‘ Coffee blossoms and insects,’ that coffee blossoming was all over 4 or 5 days ago. Apparently I found that the message sent by Mr. Jonas to the Entomologist was late, the same having been received at Coimbatore only on the day on which the blossoms were expected to open at Sidapur (viz., on the 17th April). I went to Mr. Jonas at Rock Hill where also the blossoming was all over. Mr, Jonas showed me the plants on which he was carrying on his experi- ments in that estate, Asking me to remain at Rock Hill, Mr. Jonas proceeded to Mercara to find out if there were any estates in the north Coorg where some blossoms were still expected to open, so that I might go there. “©3, Meanwhile I remained at Rock Hill making a general collection of all available insects, paying particular attention to insects visiting the coffee plants. Of the latter, there were precious few, now that the blossoms had dropped and the minute berries were beginning to set. Iam also informed by Mr. Newbery that. this year very few insects were found on cotfee even when in blossom. Stray specimens 32 of Rock Bee (Apis dorsata) and the Indian Bee (Apis indica } were found flying about in the estate, but hardly any insect of importance was found on the coffee plants. Among those found casually visiting the plants were the following :—- (1) Two species of bright metallic-coloured. Chrysomelids * (Corynodes cavifrons and Corynodes sheppardt). (2) A pinkish brown bug, Teftigoniella ferruginea. (3) The Jak Cercopid (Cosmoscarta relata).. (4) Another bug (Cletomorpha hastata). (5) Do. (Oduntopus nigricornis). “ 4. Of the above, Nos. 3 and 5 deserve some attention. The Jak Cercopid (Cosmocarta relata) is found to be a serious local pest of the jak frait tree in Coorg. The adult is a dark brown insect with reddish marks and of very active habits. The immature insects are generally found within masses of creamy white froth at the stalks of tender shoots and fruits and at the back of soft leaves. About seven or eight of these wingless nymphs are often found within the same mass of froth. In badly infested trees, these frothy masses are found in such numbers that the liquid matter dropping down often wets the soil all round a tree and there appear swarms of bees, flies and other insects attracted by the apparently sweet fluid. Many of the ripening fruits, at the stalks of which these young insects remain feeding, drop down and numbers of unripe fruits are found fallen around badly infested trees. The wingless young insects remain within this frothy mase until they assume the adult state, feeding on the sap of the tender portions of the tree ; they are dark brown in colour with light coloured bands across the abdomen; but get a reddish tinge as they approach maturity. They are found quite helpless once they are removed from this frothy matter. “5. The other insect (Odontopus nigricornis) is a Pyrrhocorid bug ; this was found in hundreds all over the estate in all stages. The bright pink coloured nymphs were found in numbers at every turn breeding especially on the fallen fruits of various kinds of trees chiefly species of Ficus, Trewia, etc. “6, I stayed at Rock Hill, Sidapur, for three days. On the 25th evening, I got a message from Mr. Jonas at Mercara asking me to go to Mercara at once. I proceeded accordingly and at the instance of Mr. Jonas, I saw Mr. Hannyngton, the Commissioner of Coorg, who told me that some blossoms were expected in Hansey Estate in north Coorg and advised me to go there. From Mercara I left for Hansey and on the way at Somwarpet I again saw the Commissioner by appointment and got a letter of introduction to the Agent at Hansey Hstate. Ireached this place early on the morning of the 29th on which day some blossoms were expected. “7. Here again, though there were some blossoms to see, the great majority of them had opened three days ago. There were a few that had opened the day previous to my reaching the estate and some that were opening on the day of my arrival. I had, therefore, blossoms of different ages to see—those that were three days old, those of the second day and some just opening. I was told by the yentleman in 33 charge and also by others in the different places I visited, that the blossoms begin to fade and gradually drop in from 36-48 hours at the most remaining for « little over two days. By the evening of the 29th, some of the blossoms which had opencd early on the morping of the previous day were noted to fade and hardly any insect was observed to visit such flowers. I think fertilization takes place very soon after the blossoms open. The best time for observing coftec-visiting insects, especially those which are likely to play a part in pollination, is, I think, the very first day. I remained at Hansey for the two days 29th and 30th, observing insects visiting old as well as new blossoms and also collecting all other available insects on the estate. 8. As in the case of several garden plants which, when in blossom, are visited by numerous insects even on the plains, I expected to come across a good many insects on coffee blossoms, but it was disappoint- ment. There were comparatively few insects visiting Huwers compared to the very rich insect fauna of the tract. The following tabular statement shows the insects noted on coffee flowers with some observa- tions on each :— 3) See Time of z Name. visitin Remarks. g the insect flow ca | belongs. 1| Hymenoptera] Apis dorsata. | Found all|‘Uhis insect (Rock Bee) was found fairly (Apidae). through | common. They are found singly generally, Bees. the day | unlike the other bee. Found collecting but pollen chiefly. Probably plays some part common | in fertilization. between 8 aM. and 2 P.M. 2 Do. Apis indica. | From early | The commonest of all insects I found visiting : morning | coffee blossoms. More than one bee is on to| found in the same flower. Extremely sunset, active both as a pollen collector and a feeder of nectar. 3 Do. Melipona .. | Common This is the insect often called in the verna- soon after | cular—the mosquito-bee. It is an in- sunrise. different feeder visiting flowers as well as rotting fruits, ete. Five or six of them 5 are often found on the same flower— chiefly feeding. 4 Do. Anthophora | Commoner | Fairly common but not as common 38 Apis zonata. during indica but commoner than dorsata. Dili- midday. gent in collecting pollen with its charae- ; teristic hum. 5| Hymenoptera] Polistes Early Several] found together, feeds on the nectar. (apide). stigma. mornivg. | The delicate paper nests of them are found asp. attached to coffee leaves.: 6) Hymenoptera) Megachile Only in| Actively colleots pollen in the pollen brush. Apidae). umbripennis els sun | Sometimes also found cutting leaves, Bee. ight. 7| Hymenoptera) Odynerus Allday .. This is a small solitary wasp often found (Wasp). fragilis. hidden inside a flower. 1 think it feeds on the nectar. It is also likely that it comes in to look for caterpillars which it has the habit of collecting and storing. 8| Hymenoptera} Ceratina All midday | Not common. Stray specimens come and (Bee). hierogly- sit on the flower, not found to remain even phica. a short time on the flower. It is a very smal] bee. 34 | Group to Time of 2 meh Name. visiting Remarks. g the insect flowers. | belongs. 9| Hymenoptera] Purevaspis All midday | Commoner than No. 8; found nibbling the (Bee). cartonaria. | andafter-| nectary. noon. ‘ 10) Hymenoptera] Prosopis sp. | {n bright] A tiny bee, pretty common, collecting (Colletide). sunlight | pollen. in the morning. : 11) Hymenoptera) Czxadro sp. At midday. | A small dark waap; nibbles the nectary. Sphegida Not common. It crawls about the flower (Wasp). stalks; probably in search of Aphida which it is known to collect and store. 12 Do. Trypoxylon In the | Fairly common ; did not note it feeding. intrudens. evening It disturbs the flower sufficiently to displace as the} the pollen grains. It is likely searching sun goes| for spiders for its nest. down. 13] Hymenoptera] Sima (Jer- | In the; A small black ant. Goes inside other (Ant). donié P). morning | flowers. Very active ; probably collects and late| food for nest. in the evening. 14| Beetles Heterorrhina | At midday. | A very active large shining green Cetoniid. Cetoniid. planata. Flies often very high. Found perching on the flowers and often nibbling the petals. While on the flower it is not watchful and is easily caught. 15} Beetles. A reddish | Early in| A fairly large leaf-beetle not very common. Chrysome- | brown the mor-| Found sitting on flower and leaf, Not lid. Chrysome- ning and| noted to feed. lid spotted late in with black.| the after- noon. 16 Do. Yellowish All day ..| Feeding on tender leaves and petals of Punyeene flowers and fairly common. 1d. 17 Do. Monolepta Early in | Found chiefly feeding on the tender leaves signata. the mor-| but few feeding on flowers. ning. 18} Beetles ,. | Klaterid Allday .. {Small black beetles found often hidden beetles. inside flowers. 19) Beotles. Luciola gor-| Karly in| These are also found partially hidden in Glow worm | fhami. the mor-| the flowers, Very likely they feed on beetle. ning. the pollen. 20 Beetles. Oleria sp... | Found in Very few found. I think it is only a casual Longicorn. the mor-| visitor without any object of feeding on ning. the flower. Not seen to disturb flower ; with mouthparts, 21| Flies fA purple | In fine | Actively visits flower after flower. Pro- blue Doli- sunshine, | bably licks the nectar, ebopid. 22) Flies .. | A dark Bom- Do. Do. do. bylid. 23) Lepidoptera. | Skipper Midday .. Extremely sa oe on the nectar ; not ie . very common. ecimen not coll h 24 Do. Sphingid .. | Late in the | Actively visiting fewer, not bad oto evening. | approached with lantern. The charac- teristic bum gives the idea that the inseot _ hawk-moth. (Specimen not ecl- ected.) 25) Rhynchota. | Meldmphaus | Noted only | A purple brown fairly large r i (Bugs.) fulvomargi- once bug, found on a ie wa Gg oe nata, apone boseis thrust into the same, only once. P.M, 35 “9. It was surprising to find that among the bees—the small honey-bee (Apis florea) was not only not seen on the flowers but I did not see a single specimen at the time of my visit. “10. There were some very big Xylocopas—the bluish Carpenter bees—flying high up and perching on trees, but I did not note any on coffee blossoms. It is unfortunate that I was not able to make obser- vations and collect, in the Estates where the experiments were arranged at the time when the blossoms opened in those places; because timely information was not received. Both Mr. Jonas and myself think that for Coorg at least, further experiments and observations should be ‘made next time to enable us to come to any definite conclusion regarding the relation of insects to coffee blossoms. “11, A good number of other insects were also collected at Hansey. On stray coffee plants, hundreds of immature hoppers of a painted variety were found actually covering the branches. These .are probably the young of the coffee locust (Aularches sp.) ; the adult insect was not seen. Probably the eggs had hatched out just after the first coffee showers. I did not find any specific pest of coffee in any of the estates I visited. “12, [left Hansey Estate on the lst May reaching Mercara the same night. I saw Mr. Jonas, who was still at Mercara and also showed him my collection the uext day. The same evening I left Mercara back for head-quarters. T. V. Ramakrisuwa AyyaR, 21st May 1918. First Assistant.” 27. The foregoing notes show that bees of various kinds do visit the coffee blossom, and that they are efficient pollinators, and it may be assumed that their presence is therefore decidedly beneficial and to be encouraged as far as possible. As regards protection of these bees, by legislation or otherwise, we are practically only concerned with one species, the Rock Bee (Apis dorsata). The Indian Bee (Apis indica) builds a nest composed of several parallel combs which are usually placed in some protected situation, such as the interior of a hollow tree; the Little Bee (Apis florea) suspends its small single comb from the branch of a bush or sometimes in a slightly sheltered place such as under the eaves of a house; the Carpenter Bee (Xylocopa) nests in dead wood, into which it tunnels so as to be destructive when the situation selected is a roof-beam or other structural timbering: none of these bees are at all gregarious. The Rock Bee (Apis dorsata), on the contrary, is markedly gregarious, its large single comb being nearly always slung from a tree or rock which also forms the home of numerous similar colonies. As a rule, the same locality is selected year after year, the bees deserting their combs during the wet season in the Hills and migrating to the plains, returning to the hills after the rainy season and building new combs in the same locality. In Coorg I was told that the bees migrate backwards and forwards twice annually, bat a aingle migration-flight in each direction seems to be the rule, that from the hills to the plains taking place about June and the return flight in November-December. Gn. . 36 28, Without a prolonged series of careful counts of the numbers of colonies ina given district it is obviously impossible to say for certain whether the bee-population (by which Apis dorsata alone is referred to) is decreasing, increasing or stationary, but residents in the planting districts are almost unanimous in asserting that it is aeashiay decreasing. As regards the causes which underlie such decrease, they may be ascribed, I think, mainly to the increased open- ing up of the planting districts concerned and consequent (1) destruc- tion of the primeval jungle and (?) increased facilities for trade. 29. In the case of those colonies of bees which nest in trees it must be remembered that there are comparatively few trees which are- suitable. One such tree may contain a hundred or more colonies of bees—one “ Bee-tree ” which I saw had about 156 colonies on it— and so many colonies require a large amount of branch-space; then again the branches must be fairly horizontal to suspend the combs properly. Before the hill districts were opened up such trees were undoubtedly much commoner than they are now. ‘The present system of shade-trees ((7revillea and such-like) over coffee does not provide the bees with suitable trees and the present ‘‘ Ree-trees” are all, or nearly ail, relics of the old jungle. As they decay or are cut down, the bees are compelled to go elsewhere and many colonies probably perish whilst searching for anew home or because they are ultimately compelled to make shift with an unsuitable position. 30. Increased facilities for trade do not. at first sight have any bearing on the destruction of bees, but, in cases where the colonies are robbed by jungle tribes, it would seem that a larger and nearer market and better prices lead to a more systematic persecution of the bees for the sake of their products. A junyle tribe which used to rob the nests for its own immediate necessities now destroys every acces- sible colony to supply the market. This is an economic factor which should not be lost sight of if the necessity for any restrictions on the robbing of these colonies of bees is considered by Government. It seems to ine that there is a decided difference between a man who collects jungle produce for his own use and one who collects it for sale to others. 31. The method of collecting the honey and wax has already been described. _The bees are always burnt off with a torch and the whole colony is killedas arnle. It has been stated (Planters’ Chronicle, VIII, 27) that “it is quite easy to take honey and wax without destroying the bees”; go it is, when the bees are in a hive under control. When they are all on a single comb containing honey and young brood, and when moreover this comb is suspended from a branch or rock perhaps a hundred feet above the groand, the matter 1s very different and J doubt whether any other method would be practicable as a rule. 32. As regards the desirability of legislation to prevent the destruction of bees, it does not appear that there is any necessity for this because the matter appears capable of being dealt with without special enactments. So far as reserved forest areas are concerned, I anderstand that the right to collect honey and wax is vested in - Government under the general heading of ‘* Minor Forest Products ” 87 and that this right is leased out annually by the Forest department to the jungle tribes who collect the honey and wax. Discoxstinuance of the issue of such licenses should form an obvious and simple check on the destruction of the bees. This applies, of course, particularly to the gregarious colonies of Apis dorsata : it is difficult to see that legislation or the non-issue of licenses would prevent in practice the destruction of single nests of Apis indica and Apis florca, as their nests are com- monly robbed when found and the honey eaten on the spot. But, in the case of the large colonies of Apis dorsata, which are always found in particular localities, which commonly store honey at particular and well-known times of the year, and for whose taking special pre- parations have to be made, there should he little difficulty in guarding against illicit interference. As has already been pointed out, the bees desert their nests at regular intervals and permission might be given to collect the wax (old combs) after the bees have flown. 33. In the case of colonies of Apés dorsata outside of reserved forest areas, they are mostly found within the limits of the estates of the planters themselves, to whose obvious interest it isto protect the bees from being harried by despoiiers and to encourage the bees as far as possible by preserving trees or localities which they have made their head-quarters. No legislation appears to be necessary to effect this : is is a matter for the good sense and co-operation of the planters who are most concerned in the matter. 34, The keeping of bees in hives has been suggested as one solution ofthe difficulty. For this Apis dorsata is unsuitable as it is too fierce and intractable, and Apis florea lives in too small colonies to be of much use, besides which the latter bee does not seern common in the plant- ing districts where the conditions evidently do not suit it; in any case, also, both these bees build only a single comb ard are therefore unsuited for life in a hive. Apis inlica, however, building several parallel combs and living normally in hollow trees and similar situ- ations, will adapt itself fairly readily to semi-domesticated conditions in the frame-hive for a short period, bunt the colonies have strongly vagrant tendencies so that they usually swarm out and fly off again after a few months. By having a dozen or more hives and constantly repopulating them with freshly collected swarms, it is possible to keep at least a few colonies going, but the trouble required is likely to be more than will be taken by any one who is not an enthusiast in the subject of bee-keeping whilst the yield of honey is too small (about 6 lbs. per hive per annum) to repay the cost and labour required for its production. This, however, takes no account of the good done by the bees in pollinating flowers—not only of coffee but of most of the flowers of the garden, shade-trees, green-manure crops, etc.—and it would certainly be advantageous to all estates if a few hives of bees could be kept. 35. General summary and recommendations.—(1) The presence of bees is not essential for the successful pollination of coffee, but (2) The natural, and therefore mos desirable, form of pollination is achieved by the aid of flower-visiting insects, of which (3) Bees are the most important. 38 (4} Of these bees, the only one which need be considered is the large rock bee (Apis dorsata). ; (5) There is fairly definite evidence that the numbers of this bee have suffered a steady reduction in the coffee districts of late years, and that : . (6) Such diminution is due to (a) to the reduction of suitable nesting-places owing to the opening up of the coffee districts and (b) to the systematic persecution of the colonies of bees in certain districts for the sake of honey and wax. : (7) It is not practicable to take the honey without destroying the bees, bat , (8) It is possible to collect the wax without injury to the bees after the latter have deserted their combs which they do at regular times of the year. (9) In districts in which destruction of these bees occurs on any considerable scale, it were well that the Forest department be requested to refuse the issue of rights to collect honey and wax in forest areas except under special restrictions as regards the number of nests to be taken or the time of year at which they may be taken. (10) Planters should regard bees as useful allies and do all possible to protect and encourage them within the limits of their own estates. (11) In any experiments in bee-keeping, the Indian bee (Apis indica) is likely to prove the most suitable. (1t may be added that the Agricultural department will always be willing to assist would-be bee-keepers with advice and help as far as possible.) T. BatnsriacE FLEercuer. AGENTS FOR THE SALE OF MADRAS GOVERNMENT PUBLICATIONS. —_>—— IN INDIA. R. Oampaar & Oo,, Calcutta. CoMBRIDGE & No,, Mount Road, Madras. E. M, GopaLakrisHyna Kone, Bookseller, Padumantapam, Madura, HIGGINBOTHAMS (Limited), Mount Road, Madras, V. Katyanarama Iyer & Co., Esplanade, Madras, G. C. Log@awaDHaM BroTHERs, Guardian Press, Madras. S. Murtuy & Co., Kuapalee Press, Madras, G. A. Natesan & Co., Madras, THE SUPERINTENDENT, Nazalin Kanun HIND Press, Allahabad, P. R, Rama Lyze & Co., Madras. D. B. TaABAPOREVALA Sons & Co., Bombay. TEMPLE & Co., Georgetown, Madras THACKER & Co. (Limited), Bombay. THACERR, Spink & Co., Calcutta, THompPson & Co,, Madras, IN ENGLAND, B. H. BLACKWELL, 50 aud 51, Broad Street, Oxford, ConstaBie & (o., 10, Orange Street, London, W.0, DEIGHTON, BELL & Co. (Limited), Cambridge. T, Fisnge UNwIin, No. 1, Adelphi Terrace, London, W.C. GRINDLAY & Co., 54, Pari t Street, London, S.W. Kxean Pavl, TRENCH, TRUBNER & Co., 68-74, Carter Lane, London, E.U. Henry §. Kine & Co., 65, Cornhill, London, B.C, ?. 8. Kine & Son, 2 and 4, Great Smith Street, Westminster, London, 8,W Luzao & Co., 46, Great Russell Street, London, W.C. B. QuaRITCR, 11, Grafton Street, London, W. W. THACKER & (o., 2, Creed Lane, London, E.C. ON THE CONTINENT ERnNzsT LEROUX, 28, Rue Bonaparte, Paris. Martinus NisHorr, The Hague, Holland.