|[ Artifical flowers in determining the winter quarters, food plants and daily movements of Lasiopticus pyrastri and other hoverflies ]
Fritz Schneider (1958)
Mitteilungen der schweizerischen entomologischen Gesellschaft 31: 1-24
A fruitful method of collection is indispensible to field investigations of activity, movement and feeding of hoverflies (Syrphidae). As likely places for collection we consider places preferred by males for seeking out females; drinking places after a long dry period; plants attacked by aphids where the flies lay eggs or lick honeydew; and finally flowering nectar- or pollen-plants. In spring, during the major activity period for predatory hoverflies, there is a wide variety of flowering plants and trees available, and the animals are distributed over a very wide area. In these circumstances it is often not easy to collect many flies of the same species. In summer and autumn conditions are rather more favourable.
As a result of the observation that hoverflies are sometimes attracted by yellow clothing or pieces of paper, we developed a new collecting method in 1953, which is very well suited to the study of certain problems (Schneider 1954). Large yellow paper flowers which attract hoverflies and other insects, particularly in December, or at the end of winter in February and March, when overwintered flies are empty of food and must consume large quantities of nectar and pollen for the development of their ovaries, while as yet only isolated flowering plants are available. Lasiopticus pyrastri belongs to the group of predatory hoverflies which overwinters as adults, and it responds particularly well to our method. As a destroyer of aphids this species is a familiar 'gardeners friend' in agriculture, horticulture and fruit-growing, and it will often be effective very early on, before the commencement of aphid reproduction.
I owe particular thanks to Dr Maurizio ......
II. Method of collection
The type of paper flower that we use now (Fig 1) is the product of past experiments and practical application, and has proved itself well. However it is of course possible that equally good results could be achieved using different shapes and sizes of flower. We also took care to choose rough terrain and a sharply contrasting background, in order to widen the area of attraction as far as possible.
[ Fig 1 shows two groups of artificial flowers in the collecting spot above Amden, and two individual flowers ]
The flowers are made of strong white paper on thin card, in the shape of a 12-point star some 15 cms in diameter. They are sprayed with an indissoluble scentless luminous light yellow colour (e.g. Zaponechtgelb made by BASF - i.e. true yellow zapon), and after drying are painted over with a thin colourless sugar glaze. In the middle a round hole of 7 mm diameter is punched. They are then stuck to squares of black cardboard, and treated on the undersides with paraffin wax: this ensures a firm hold on the ground for the display, heightens the contrast effect, and protects them against moisture from the ground. Between the flower and the card one slips a strip of blotting paper, onto which a drop of scent is introduced via the round central opening immediately before use. As this scented 'bait' we used equal parts of peppermint oil, thymol and oil of aniseed.
The edges of woods on steep south-facing slopes are particularly suitable collecting places. The flowers are laid out in the morning of a stilll sunny day in isolated groups of 10-25 around a knoll or undulation in the ground. They are placed such a way as to be visiabl not only from above, but also from the side from various aerial positions. On a calm day the flower fields are enveloped in a cloud of scent. Provided wind and temperature conditions are favourable, the approach of flies from the surrounding area is usually immediate. The flies settle on the sugar layer, which they begin to dissolve with their saliva. This method of taking food is, however, much more troublesome and takes considerably longer than is the case when food is offered in liquid form. There is consequently plenty of time to catch the flies on the isolated groups of flowers with a net. Sometimes the flowers exert such an attraction for Lasiopticus pyrastri that one can bundle them straight into collecting bottles without the use of a net: those that escape do fly away, but settle immediately on the next flower. During a really heavy influx many more syrphids land on the flowers than can be caught by one person.
III. Collecting places
In order to avoid repetition, the collecting sites of our specimens are here briefly described and numbered.
1 Durchschl„gi above Amden (Canton St Gall): 1160 m. Stony, grassy slopes, falling away steeply to the S, SW and SE. Large 'catchment area' in front of the edge of a spruce wood. Best collecting place so far for L. pyrastri.
2 Weesen (Canton St Gall): 600 m. Meadow, surrounded by scrubby woodland and undergrowth, falling away steeply to SE. Village to NE. P653 to SE [?]
3 Etzel (Canton Schwyz). SW slope, 980 m. Meadow in front of spruce wood.
4 W„denswil-Gerenholz (Canton Zrich). 470 m. Meadow adjoining gardens falling away to NE.
5 W„denswil-Dorf (Canton Zrich), 630 m. Meadow in front of mixed woodland, falling away to SW.
6 Marthalen (Canton Zrich), 410m. Mixed oak-beech wood to SE of village; even ground in a small clearing near the edge of the wood.
In what follows, bracketed numbers refer to these sites.
IV. Insects other than syrphids which were attracted
As a rule we restricted the scope of our collection to hoverflies: we therefore only made isolated sporadic observations of other insects. However, the method is suited to the collection of members of other fly families, and often these were so numerous that they had to be driven away - otherwise they would soon have finished up the sugar layer altogether.
In the vicinity of apiaries, bees were regularly attracted in February and March, although they can also be caught in large numbers on Carex pollen near paper flower groups, not to mention the arrays themselves. On 17.xii.53 (1) we observed bumblebees, several Small Tortoiseshells (Aglais urticae) were seen, and Vanessa io (Peacock), Polygonia c-album (Comma) on 13.xii.57. Flies other than syrphids appeared only singly on 13.xii.53 (3) and on 17.xii.53 (1); they were more numerous on 24.ii.54 (4) and 10.iii.54 (5). On 11.iii.54 (1) there were 5-10 per flower, and on 22.iii.54 (3) there were 20-30 per flower, so that from then on they had to be driven away. Bees, bumblebees and butterflies normally make for the central round hole in the paper flowers and attempt to suck nectar from here as in normal flowers; whereas flies, including hoverflies, alight on any spot on the yellow paper flower, and commence feeding immediately - as soon as the feet touch the sugar layer.
The following list contains fly species collected as a random sample in four places over six days. M=male W=female. Bracketted figures refer to collecting sites; unbracketted figures refer to the nukmber of individuals and the dates [list].
V. Syrphid catches
In table 1, some of the most fruitful syrphid catches are set out. As a rule the 'bait' experiments were restricted to the period from December to early April when there are few flowers. On sunny cloudless and windless days L. pyrastri started flying on occasion in shade tempreatures as low as 4-5 C. On 25.xi.53 on the Etzel, three flies approached between 1500-1530 at 5 C, although none alighted. On 23.ii.54, however, when the temperature rose from 2 to 5 C between 1230 and 1450 there was not a syrphid in sight. In Amden L.pyrastri started flying on 11.iii.54 in still sunny weather, at 1035 at a shade temperature of 4 C. At 1200 at % C there came a mass approach, whilst later on at 1430 the flight was cut short when the sky clouded over. As well as shade temperature, cloud cover has also to be considered. At 7-9 C flight is frequently lively but can be depressed again by an increase in cloud cover (Amden 17.xii.53).
The attracting effect of the yellow paper flowers is not 100% effective before the overwintering period in November-December. On the Etzel (25.xi.53, 13.xii.53) and in Amden (17.xii.53), Lasiopticus individuals approached the array, hovered over it, but then passed on without being affected by it. They seemed to respond weakly to flowers, but, by contrast, many of this species flew next to spruce and fir trees, many of them settling on twigs and dabbing at the needles with their proboscis. At this period, there seems to be a definite preference for honeydew, to be stored in the crop as a winter supply. In winter, then, as with the maturation feeding in early spring, they change their minds about flower visiting.
In March, Lasiopticus seem to notice the yeloow paper flowers from a great distance. The flies often land at great speed, spiralling or rushing down on the flowers and starting to feed immediately. In two cases the outsize paper flowers attracted only L.pyrastri females from a distance. Then the flies shifted to a familiar flower type, between the paper flowers, and settled on Carex or Bellis. In the second case the fly was caught immediately: the crop contents consisted entirely of Bellis pollen, which points to a definite learning ability, and a more than moderate flower constancy.
Deep lemon-yellow flowers have proved effective in catching L.pyrastri. On 13.xii.53 on the Etzel they were clearly preferred to orange, red, lilac, and ultramarine ones, and also to light lemon-yellow on the 17.xii.53 at Amden. Here too, light red, light mauve, and ultramarine did no better, although occasionally a blue one had more visitors. Even isolated scentless yellow flowers will attract L.pyrastri (Etzel 22.iii.57).
On 17.xii.53 in Amden, in a group of yellow flowers, certain inclining flowers at the edge of the array, and set at a sharp angle, were clearly preferred.
Four experiments were carried out in June with no success, even in places where syrphids were known to occur. Probably this was due to the combined effects of the abundance of food (flowers, honeydew) and of greater constancy of place. Also, it is not generally possible to entice males away with paper flower from their mate-searching flight in clearings of woods. [ TABLE 1]
VI. Overwintering of syrphids as adults
Most predatory hoverflies overwinter in a late larval stage; however it has already been possible to prove overwintering in an adult state for L.pyrastri, L.seleniticus and E.balteatus (Scheneider 1947, 1948).
This unusual behaviour can be demonstrated if it is possible to cpature flies with a full crop and well-develo0ped fat body before hibernation (L.seleniticus), to find flies in their hibernation sites, or to collect mated females on their first flight when there are no males in early spring. Table 1 confirms our earlier findings:
E.balteatus: in the samples from 13.xii.53, before overwintering, males still appeared, but in the spring samples only females were found. All these females were dissected and their spermathecae were without exception full of sperm. Their ovaries were as yet undeveloped (0.9 x 0.5 mm to 1.3 x 0.65 mm); only flies in the sample of 15.iii.57 had eggs which were already mature.
L.pyrastri: this species is by far the most numerous in our sampling programme. Before overwintering the proportion of males in both samples was about 30%: after overwintering it was 0% in five samples, and 3.6% and 5% in two others. A number of females from all these samples were dissected - 59 altogether. Apart from two uncertain cases, all had full spermathecae. From all the samples, 133 females were investigated for the developmental state of their ovaries. In every case the ovaries were as yet undeveloped: 1 x 0.8 mm - 1.9 x 1.1 mm, or 1.8 x 0.8 - 2.6 x 1.2 mm. This was different only in samples taken on 13.iii.57 and 15.iii.57, when 35 or a sample of 39 females had 1/2 to 2/3 developed ovaries. These findings show again that females caught in March and April must have overwintered, and not emerged in spring. Pollen analysis points to a similar conclusion, and this will be dealt with later. As well as typical spring poelln types, many flies contain the remains of summer and autumn flora from the previous year. The only new finding is that evidently, under certain conditions, isolated males are also able to overwinter.
L.seleniticus: whilst prior to overwintering males are more common than females, all five samples after overwintering (in which seleniticus appears at all) show a complete absence of males. All these flies were dissected, and contained sperm in their spermathecae. All ovaries were undeveloped (1.2 x 0.7 mm - 1.5 x 1.0 mm), excepting one isolated case taken on 13.iii.57 which had a 3/4 developed ovary. These results too point to overwintering by mated females.
Metasyrphus lapponicus: In our samples this fly did not appear before overwintering, and only occurred four times afterwards, all cases being females. All contained full spermathecae and undeveloped ovaries (1.3 x 0.5 mm - 1.9 x 0.9 mm) excepting one taken on 22.iii.54 with an empty receptaculum seminis. This species can therefore also be grouped with those species that overwinter as mated females.
In 1957 in Weesen/Amden, L.pyrastri and M.lapponicus were even less advanced than E.balteatus in ovarial development. The body size of overwintered females shows a marked variation in L.pyrastri (...data...). Undersized specimens of all three species mate normally, and survive the winter.
The pattern of occurrence of Eristalis tenax and Cheilosia melanopa indicates likewise that they overwinter as females, although analysis of spermathecae is not available in these species. It is not possible to deduce adult overwintering from the data available for Metasyrphus luniger or M.corollae. For Melangyna lasiophthalma, M.quadrimaculata and Cheilosia insignis, we are speaking of protandrous species, whose males emerge very early in spring, before females.
The following method has proved effective in showing the presence of sperm in the spermathecae of syrphids. The three shperical black receptacula seminis are immediately removed once the animal has been killed with ether, and placed in physiological saline solution without the surrounding colourless gelatinous layer. They are then transferred to a slide in a drop of physiological saline solution and carefully covered with a coverslip. They are then quickly placed under a microscope, where light pressure is exerted on the coverslip until the spermathecae burst, allowing the sperm packets to come out. With a little practice the technique becomes easy and useful.
VII. Pattern of syrphid approaches to paper flowers
Fig. 2 shows the pattern of a fairly consistent approach by both sexes of L.pyrastri on 17.xii.53 above Amden. Approach flight begins at a shade temperature of 7 C at 1030 and steadily increases with the temperature until 1230. At 1230 there is a sudden rush of male L.seleniticus. Between 1000 & 1300 the approach of S.luniger is completely constant.
[ Fig. 2: continuous approaches made by L.pyrastri on 17.xii.53) above Amden. Points = females, Circles = males. Ls = L.seleniticus, Sc = Metasyrphus corollae, Sl = M.luniger ].
Observation of flight on 22.iii.54 on the Etzel (Fig 3) had to be cut short at 1600. Approaches made by L.pyrastri are of an amazing intensity and constancy. If only the local syrphid population was attracted by the flower array, the approach would increase with time. However, there seems to be a constant circulating population extending over great areas of meadowland and spruce forest in the upper valley of Einsiedeln. In terms of numbers, the species L.seleniticus , Metasyrphus lapponicus, and E.balteatus are far outnumbered.
[ Fig 3. Constant approach flight of L.pyrastri on 22.iii.54 on the Etzel. ]
In W„denswil-Geren on 14.iii.54 (Fig 4), following the massed approach of L.pyrastri after 1200, there was a long pause around 1300 which cannot be explained by weather conditions: the temperature is rising constantly and the weather remains sunny, cloudless and still. During this hour there seem to be few flies circulating in the neighbourhood of the collecting site. At 1330 a strong and continuous approach begins, which is interrupted at 1440 by a sudden increase in cloud cover. Here, too, the three other species are poorly represented.
[ Fig 4. Sporadic approach of L.pyrastri on 14.iii.54 in W„denswil-Geren ]
There is a similar interruption in the appraoch flight in Fig 5 (Amden 11.iii.54). After the first peak of L.pyrastri between 1040 and 1100 at a shade temperature of 4-5 C, there is a pause [P] despite ideal flying conditions (cloudless sky, practically no wind). Between 1100 and 1130 only isolated individuals seem to be present round the collecting sites. At 1130 the approach resumes and by 1200 grows so massive that it is impossible to collect all the flies at once. [M = mass approach] When the attraction effect is at its height, the flies will allow themselves to be collected by hand, without attempting to escape. On each flower there are perhaps 5-10 other flies. With increasing cloud and gathering wind, the approach recendes after 1340 and remins weak til 1500.
The unusual flight pattern from 1000 to 1330 cannot, in this instance, be explained away by cloud cover and temperature. After the first flight over the capture site there is a period where the flies go somewhere else - perhaps foraging in the valley - before returning to the higher regions 1-2 hrs before the cloud starts to come across. The area surrounding the study site is thickly covered with snow as far down as the centre of the village (Amden) - 1160 to 900 m - and the flowers are arrayed ona small, steep patch, free of snow. Nothing is in flower until some 700 m - further down. Crop analysis and pollen preparations have shown (see below) that the flies do spend time in the valley, if only periodically.
[ Fig 5. Sporadic approach of L.pyrastri on 11.iii.54 above Amden ]
On 13.iii.57 we were collecting continuously above Amden from sunrise to sunset, using 100 paper flowers. The prefumed yellow flowers were set out in seven isolated groups. Approaching flies were transferred individually to glass tubes, stating the time caught. The material was labelled at 1/4-hrly intervals and transferred to a cool dark place. We determined shade temperatures and wind speeds (with an anemometer) at similar intervals.
The weather was mild and sunny all day. The whole trough around Amden, extending far above the study site, was already snow-free (snow-line about 1600 m). The approach curve of L.pyrastri shows the following pattern (Fig 6):
[ Fig 6. Respective approach curves of L.pyrastri in Amden and Weesen on 13 and 15.iii.57, which suggest daily wandering. Temperature and wind strength: straight line = Amden; dotted line = Weesen. Chilos. = Cheilosia insignis ]
Flight begins as early as 0800 at a shade temperature of 6 C, but remains strikingly weak until 0900, when a definite peak is observed, dropping off around 1100. The creatures swarm down from the heights to the flower groups, but we have been unable to observe where they come from.
Between 1045 and 1500 all is quiet around the artificial arrays, aprt from two waves of approach at 1245 and 1415 which soon die away. The, after 1500, a mass flight begins which lasts uninterrupted, wells up towards 1645, and the suddenly breaks off, stopping altogether after 1730. At 1630 we observed three times little groups of L.pyrastri flying up from the valley from a SSE direction, close to the ground, towards the edge of the wood. Most of the insect alighted on the fields of flowers, but others flew towards the conifer woods, some hovering over the flowers for a short time. The flight curve shows once again a pattern of occurrence very similar to that of 14.iii.54 and 11.iii.54: a deep saddle, inexplicable by temperature, cloud cover or wind strength. The peak at 1600 probably coincides with a drop in wind speed from about 2 to 1 m per sec; but the smaller peaks at 0915, 1245 and 1415 show the same wind speeds as the flight minim at 1200 and 1330. Besides which, a wind of 2 m per sec (7 Km per hr) is far too weak to have any marked effect of the flight of L.pyrastri, let alone prevent it.
The consistent flight of Cheilosia insignis adds further weight to the argument that weather conditions do not explain the weak flight of L.pyrastri in the hours around noon. It is far more likely that in the morning and the evening, the study sites are passed over by loosely-knit swarms of L.pyrastri, and that in the meantime the flies go somewhere else. Direct observation proves, not to mention the crop analyses to be considered later, that towards the evening the flies return from the valley to the mountain forests.
However, if L.pyrastri does fly down from the mountain forests of the Mattstock into the pollen- and nectar-rich meadows in the valley, returning at evening to their high-lying night quarters, then the peak period of flight in the valley must occur around midday. We therefore repeated the experiment two days later above Weesen (600 m, i.e. 560 m lower than the study sites above Amden) and were able to confirm our assumptions. The weather conditions were very similar to those in Amden previously; the weather was mild and sunny (dotted line in Fig 6). Flight began at 0830, peaked at 1200, grew sporadic by 1400, and died out by 1600 - i.e. by the very time when, two days beforehand, the return flight from the valley had reached its peak. The total number of flies approaching was much smaller than in Amden, as we were using fewer flowers, and also because over the wide area of the meadows the flies naturally spread out far more than in the concentrically approached site above Amden.
Considered separately, the flight curve for Amden could be seen as an indication of an activity pattern which changes according to the time of day, without any essential change of locality - although it is rather implausible that L.pyrastri should spend the part of a March day with the best weather conditions (relative to sun and temperature) inactive in some hiding place. The complementary timing of the appearance of L.pyrastri in the two sites shows, however, that at this time of year the flies do not so much change their activity as their locality during the course of the day, and that they undertake daily wanderings over sonsiderable distances. Only the two ends of the Amden flight curve, which are explicable by the leaving and returning to the night quarters, are activity determined.
[ Fig 7: Area of daily wandering of L.pyrastri between Gschwensh”hi (Amden) and Weesen.
A = Amden collecting site. W = Weesen site. N = coniferous forest on the Gschwendsh”hi (1160-1400 m) winter- and night-quarters of L.pyrastri. L = deciduous woodland with Anemone hepatica, Viscum, etc ]
Fig 7 shows the topographical conditions. The spruce wood (N) (1160-1400 m) is the preferred overwintering and night quarters for . For phenological reasons the most favourable feeding grounds after overwintering lie in the valley below (L & W). The insect is an excellent flyer and copes easily with these distances of 1-2 Km, and differences in altitude of about 700 m.
I seems that in W„denswil-Geren (Fig 4) similar conditions to those in Amden prevail. The little Geren wood with its stands of spruce can probably be seen as night quarters. From here - in contrast to Amden - the flies can spread out in all directions because of the isolated position of the little wood. The high flight concentration at 1200 corresponds to the exit from the night quarters, and the second flight maximum at 1600 to the return. Further experiments must be done to show whether it is possible to consider as night quarters for flies in the W„denswil area the conifer-covered slopes of the Hohen Ron, some 5 Km away.
Even the intensive and excpetionally consistent approach flight on the Etzel (Fig 3) can be fitted well into this frame. The upper valley of Einsiedeln (900 m) is surrounded by a chain of wooded mountains. The spruce woods are not far from the valley floor. The night quarters are numerous, and in terms of height are scarcely separated from the feeding grounds; however, as a result of this lack of phenological differentiation in the area, there are often food shortages at the time of the first flight, which prompt the flies to extend their flight on the horizontal rather than the vertical plane. This flight behaviour leads automatically, however, to a wide-ranging circulation and hence to the consistently constant approach which we established on 22.iii.54. The continuous flight on 17.xii.53 above Amden is equally understandable in that, at that time of year, it is impossible to think in terms of a phenological division of food plants based upon altitude.
One further observation is worthy of mention. The massive return flight to the night quarters takes place at quite different times. In Amden on 13.iii.57 at about 1600; in the same place on 11.iii.54 as early as 1200 and 1300; in W„denswil-Geren at 1400. In every case this concentrated return flight takes place n mild sunny weather. If we now consider schematically the factor of cloud cover in the flight diagrams of 11.iii.54 and 14.iii.54, it is striking that the sky quickly clouds over as soon as the massive return flight is finished. The flies occupy their night quarters immediately before the weather deteriorates. This cannot simply be a question of the flies immobilising on the spot as a result of the cloud cover, for in W„denswil-Geren the flight remains weak even later on, when the cloud completekly disappears again after 1500. The assumption that the cloud simply interrupts an otherwise normal flight is also countered by the extraordinarily stormy flight in Amden between 1200 and 1300 (11.iii.54). It would be tempting to investigate this question further experimentally, to see whether L.pyrastri actually does rely on a sense of the meteorological changes which herald this increase in cloud cover. Such a faculty could prevent flies being overtaken by a storm at the end of winter, before they had reached suitable night or overwintering quarters.
Unfortunately we have not so far been able to discover L.pyrastri in its winter hiding places. On the site at Amden we could merely observe how, on 13.iii.57, flies of this species htronged upwards to the mountain spruce woods. On 29.i.58, in sunny weather, many L.pyrastri and L.seleniticus showed themselves around midday on the same site, at the edge of the wood and inside it. At 1350 - 1420 at 5 C, several flew close to the ground upwards towards the mountains and into the wood. Two individuals sat down and showed some interest in the folds of a dark-brown dress and in the inside of tufts of grass. It appeared that they were looking for hiding places on or in the ground.
VIII. Syrphid feeding
Theoretically we have at our disposal several methods of investigating the feeding of adult syrphids. One of the most obvious would have been direct observation of the fly on its food plant: however, syrphid populations (and in most cases the food plants) are distributed over so wide an area that even with a considerable amount of work, few individual observations could be recorded. In addition, this method has the disadvantage of emphasizing those food plants which are easily accessible to the experimenter, at the expense of food plants which are preferred by hoverflies but are difficult to reach. The paper flowers offer amore impartial view, because they are new to all the flies, irrespective of their earlier feeding, and thanks to their great powers of attraction they offer a relatively effortless and very rich source of experimental material. A previous flower visit can be shown from pollen. Usually single pollen grains are caught in the flies' hairs, though not in quantities sufficient for our experiments. Pollen in the excrement is often so far broken down and modified as to make positive identification difficult or impossible.
Crop analysis produced the best results. With flies with undeveloped ovaries, one can proceed as follows. The flies are killed in ether; then, using fine watchmaker's forceps the first abdominal segment is grasped from the side across its whole breadth and separated from the thorax. Now the abdomen, opening forwards, is laid on a broad coverslip and the contents pressed out onto a slide. After each operation the forceps are carefully cleaned even if they have not been in direct contact with the crop contents. The dried drop on the slide is then sealed with warm glycerine-gelatine.
The time at which pollen was taken and its physiological significance can vary greatly. Before overwintering the flies fill their crop predominantly with liquid containing sugar - nectar, and above all, honeydew. This however does not prevent a few isolated pollen grains from summer and autumn remaining in the crop, and getting mixed up in spring with the pollen from early-flowering plants. Single grains also find their way into the crop along with nectar in spring, without this indicating the onset of maturation feeding. Only when large quantities of the same type of pollen are found in the preparation, or a great deal of pollen from flowers which produce no nectar, can it be assumed that the fly has eaten the pollen directly from the anthers, and has begun its maturation feeding.
The samples taken on 13.xii.53 and 17.xii.53 consisted of the remains of the autmn population which could not survive the winter. The crops of these flies were empty or almost empty; in contrast to overwintering individuals, many L.pyrastri and M.luniger contained even greater quantities of pollen. Kept in the dark at 5 C, they all died within 1-3 weeks, males first, then females.
Flies from the first catches in spring rarely contained pollen. On 10.iii.54 (W„denswil), of 31 L.pyrastri females, we only found pollen in four cases (Corylus, Galanthus, Bellis, Viscum); in three L.seleniticus females only one case (Corylus). But on 14.iii.54 (W„denswil-Geren) of 13 L.pyrastri, 11 had eaten pollen; of 3 L.seleniticus two had eaten pollen; of 4 M.lapponicus 3; although of 5 E.balteatus only one. Corylus and Bellis were once again the most frequently visited plants. Later, the proportion of pollen carrying flies increases still further. On 13.iii.57 127 of 138 L.pyrastri were already carrying pollen in their crop.
Table 2 shows which pollen types can most frequently be found. shortly after leaving the overwintering site, in the four syrphids L.pyrastri, L.seleniticus, M.lapponicus and E.balteatus. The values can hardly be standardized, but refer to the given localities or places with a simlar flora. Of the woody plants, Corylus is dominant among the early catches: hazel pollen is evidently very popular. It is also very suitable for lab cultures, both for feeding and for stimulating ovarial development. Ulmus, Salix and Alnus are also well visited, but Fraxinus and Populsu more rarely.
It is surprisingly common to find the giant pollen of mistletoe (Viscum album) in L.pyrastri, L.seleniticus and M.lapponicus. This parasitic plant is quite often found around the Weesen area on fruit and forest trees. It flowers early, at the beginning of March, and on account of its readily accessible male and female flowers, produces large quantities of nectar. Mistletoe blossom is very popular with Diptera, and seems to represent an important first food source for those species that overwinter as adults. On 15.iii.57 many mistletoe plants were attacked by psyllids (Psylla visci), whose nymphs and larvae produced large amounts of honeydew.
Anemone and Bellis dominate the herbaceous plants. In Weesen and Amden we are speaking in the first place of Anemone hepatica, which is very common in the deciduous woods above the Walensee. Ranunculus usually means R.ficaria. As well as the species mentioned in the Table, the following could also be found in L.pyrastri: Cirsium, Colchicum, Gentiane, Lamium, Luzula, Petasites, Potentilla, Taraxacum, Thuja. Occasionally fungal spores are also found in the crop. In M.lapponicus, Iris pumila is also noteworthy. Of the more unusual pollen types, various ones will have come from the previous year.
The menu of L.pyrastri and other syrphids is therefore pretty varied. Almost all early flowering plants are visited, with a few exceptions (eg Primula, Viola, very common on 3.iv.56 in Alpnach) which are perhaps not easily accessible for Diptera. It is not possible to point to one preferred flower shape or feeding area (eg on the ground or in the trees); in this respect, L.pyrastri has multiple tastes. Only concerning colour: yellow seems to predominate among all these food plants - ie the colour that makes our paper flowers so attractive. However, white flowers (Galanthus, Bellis) were not ignored, and neither were blue ones (Anemone hepatica). The proven vists of syrphid species to naturally and artificially yellow-coloured flowers tally well with the sensory physiological experiments on Eristalis tenax reported by Kugler (1955).
[ Table 2: relative frequency of various pollen types in the crop of female syrphids in March-April. Figures without brackets: number of flies in which relevant pollen was proved. Bracketted figures: no. in which large quantities were present (maturation feeding) ]
IX. Pollen combinations in individual syrphids
The fact that syrphids with pollen in their crop will land on the oversized paper flowers (which do not occur otherwise in nature) shows that they are not fixedly attached to one particular flower type. Neither their spontaneous reaction nor their learning capability prevents their visiting a flower type which cannot be rooted in their innate behaviour, nor in their individual experience. Evidently those characteristics of the paper flowers which distinguish them from natural food plants are of secondary importance to syrphids when set against the shining yellow colour and flower shape. It was therefore a reasonable assumption that an individual syrphids would display equally little flower constancy towards its natural food plants.
In Table 3 are collated the most frequently occurring pollen combinations in individual L.pyrastri. Insects with just one pollen type (column 2) are less frequent than those with two or four different pollen forms. Only one fly has visited only Viscum, 13 visited one more type besides, 8 visited two further, and 10 three further type. With Anemone too, insects which have only visited this type are no more common than those which have visited one or two other plants as well. The generality is the successive exploitation of a variety of different flower types. Thus one not only finds combinations of woody - or herbaceous - plant forms, but frequently also combinations of wwody with herbaceous pollens, such as Corylus with Anemone, or Salix with Bellis.
In isolated cases we found five to eight different pollen types combined in the crop of L.pyrastri. [list]
Certain isolated pollen types probably come from the previous year. Some spring flowerers seem to appear at staggered intervals, and force the syrphids gradually to change. Nonetheless there remain many simultaneously flowering species whose pollen found in combination in the same crop contents points to a broad spectrum of behaviour. We have already mentioned one exmaple of flower constancy (preference of Bellis to paper flowers following an earlier Bellis visit), but in practice it seems that such a fixation is of subordinate importance. It seems far more likely that in early spring the widest possibilities of insect feeding are exploited, and the food-plant changed to the circumstance and opportunity.
[ Table 3: Type and frequency of pollen combinations in crop of L.pyrastri females from various sites, captured 10 March - 11 April ]
X. Pollen analysis as evidence of wanderings
One may adduce that a hoverfly has been wnadering if many pollen types found in the crop have been released by plants which have either not yet come into flower in the greater environment of the study site, or which do not occur there at all. However, pollen analysis tells us little about the length of time necessary for these changes of location, or whether a periodic daily wandering takes place. The first indication of greater flight distances was provided by the catch of 11.iii.54 in Amden. That day theref were no flowers out within a wide area of the study site. In the valley below at Weesen, Bellis, Anemone hepatica and some garden flowers were in flower; Corylus was already over. The slope was free of snow up to about 650 m; from 650 - 900 m there was a patchy covering of snow; above 900 m was comprehensively covered with old snow. The flowers were arrayed on the site (1160 m) on a little knoll. Practically nothing was flowering above 700 m, so syrphids bearing pollen must have flown up from the valley. Of 29 L.pyrastri examined, 19 contained pollen (see Table 2); of ten L.seleniticus, five. Two flies had already begun their maturation feeding on Corylus, Viscum, Anemone, Bellis, Crocus and Alnus.
Fig 8 is a graphical representation of the relative frequency of various pollen types in the flies examined from Amden (13.iii.57) and Weesen (15.iii.57). It must be remembered that the total from Amden does not refer to the total number of flies (L.pyrastri) that landed or approached, but simply a number of deliberately chosen samples from the day's flight.
[ Fig 8: Frequency of different types of pollen in the crop of L.pyrastri on 13.iii.57 and 15.iii.57 in Amden and Weesen ]
In Weesen however, every fly that arrived was dissected. In connection with the wanderings of L.pyrastri mentioned earlier between Amden and Weesen over an increase in altitude of 700 m and a distance of 1-2 Km, it will be interesting to compare the pollen content of flies from both sites (i.e. the toal from Amden and Weesen). Common to both sites is the great frequency of Anemone, Bellis, Corylus, Ulmus, Viscum and Salix, although in Weesen Anemone and Corylus are somewhat ousted from prominence by Ulmus and Bellis. Nonetheless, from these and similar insignkificant differences one would never deduce a difference in altitude of 600 m at this time of year. If we now look also at the individual samples taken at different times of day in Amden, we see that the differences which separate these individuals are at least as great if not greater than those between the total numbers from Weesen and Amden. The pollen content of flies from both sites is, then, practically identical and shows again that the flies move from the site of Amden which is poor in flowers to the lower-lying feeding places, and return again from there to the upper regions.
Between 0800 and 0900, individuals land in Amden which have a strikingly high proportion of Corylus pollen, and few pollen types from the previous year. These are probably insects which leave their wintering sites very early, and have thus been able to exploit the hazel flowers in the valley. The later sections (see Fig 6) contained Anemone and Ulmus pollen rather than Corylus, and alos had far more pollen from the previous year, which would indicate a later commencement of feeding. Also noteworthy is the large proportion of Anemone pollen at 1600 - 1645, i.e. at the same time at whcih small isolated groups of flies were flying from a SSE direction up towards the wood. If we follow this flight path back, we end up on the deciduous-wood covered precipice which gives from Amden through onto Lake Walen, and in whcih A.hepatica is very common. The difference in pollen content of the flies colloected at various times, and the often sporadic and broken approach flight to the paper flowers is therefore probably explicable thus: the flies leave their night quarters at various times dependeing on their previous history, and return to their night quarters at various times depending upon their feeding sites.
A type of large yellow paper flower suitable for the catching of syrphids and other Diptera is described. Apart from syrphids, members of the Muscidae, Tachinidae, Sepsidae and Helomyzidae are chiefly attracted. The attracting power of the paper flowers on syrphids reaches its peak in March, i.e. when there are few food sources available and the flies are programmed for pollen feeding for the development of their ovaries. Our earlier findings, that L.pyrastri, L.seleniticus and E.balteatus overwinter as mated females were once again confirmed, and M.lapponicus added to the group. Occasionally individual male L.pyrastri also manage to survive the winter.
The approach of L.pyrastri to the paper flowers in the course of a day can be separated, according to collection site and season, into a continuous or a sporadic approach. We were able to show that the flies prefer spruce woods as winter quarters, even if they are far away from their feeding sites. The sporadic nature of approaches to the artificial flowers is often connected with this wandering between night quarters and feeding sites. The flight curves for valley and mountain sites are therefore not equal but complementary. The flies return to their night quarters prior to a massive increase in cloud cover.
Pollen analysis gives a key to syrphid feeding. In the crop of L.pyrastri on March 28 different pollen types were found. Important food plants (for olther early-flying syrphids as well) are Corylus, Ulmus, Salix, Viscum, Anemone and Bellis. Some isolated pollen forms come from the previous year. From a tabular representation it can be seen that folies with several pollen types in their crop are more common than those with only one. They exploit almost every possibility open to feeding flies, change food frequently, and show very little tendency to flower constancy. Pollen analysis too indicates a wide ranging daily wandering by L.pyrastri. In mountain sites before the snows melt, flies contain pollen from flowers which are only in flower in the valley. On 13.iii.57 and 15.iii.57 the pollen spectra in the crop of these syrphids from Amden and Weesen were almost identical, despite a 600 m difference in altitude.
Translated by Hilary Gilbert, 1980