Bumblebees of the upper Brembo valley
(preliminary version - 6 November 2010)
The distribution of Italian bumblebees is very imperfectly known, as is apparent, for instance, from . This is the case even for a region such as Lombardy, which has a long history of entomological exploration. For example, there are large portions of the alpine part of the region for which data are few or completely lacking. It is therefore of some interest to study in detail the bumblebee faunas of specific parts of the region. This has for instance been done by Intoppa et al. (see , but also ) for the upper portion of the Val Camonica (the valley of the Oglio river). In this note I will attempt to describe, based on field work carried out during the years 2001-10, the bumblebee fauna of a rather restricted portion of the Orobic Alps north of Bergamo straddling the divide between the basins of the rivers Brembo and Adda, and lying almost entirely to the south of the divide, in the province of Bergamo.
1. The study area. The area explored (maps) encompasses approximately 20 square kilometres in the upper reaches of the Val Brembana (the valley of the Brembo river, in the province of Bergamo) and in some adjacent parts of the province of Sondrio. It lies almost entirely in the drainage of the Mezzoldo fork of the Brembo river (Brembo di Mezzoldo). Minor parts of it belong to the drainage of the Valleve fork of the Brembo (Brembo di Valleve), to the Valli del Bitto (Valle del Bitto di Gerola and Valle del Bitto di Albaredo), or to the Val Tartano. The area comprises parts of the territories of the municipalities of Valleve, Piazzatorre, Mezzoldo, Averara, Santa Brigida, Cusio and Ornica in the province of Bergamo, and of Gerola Alta, Albaredo per San Marco and Tartano in the province of Sondrio. It includes large portions of three mountain groups, namely the eastern end of the Pizzo dei Tre Signori group, east of Cima Piazzotti, the Mt. Ponteranica group, and the Mt. Fioraro-Mt. Azzaredo group all the way to the San Simone pass. Elevation varies between 740 and 2378 metres above sea level, and most of the area investigated lies above 1200 metres. The climate, typical of the outer ranges of the central Italian Alps, is sub-oceanic and rather more humid than the one of the inner alpine ranges. Rainfall is abundant (almost 2 m annually at higher elevations), with a marked summer maximum, while summer nebulosity is considerably higher than in the inner Alps. Rocks are predominantly acidic; the study area adjoins a major limestone outcropping, the Cavallo-Pegherolo mountain group, but overlaps with it only marginally. The tree-line is generally situated at an elevation of approximately 1700-1900 metres. Below it, most of the land is densely wooded, except, of course, where the forest has been cleared to create hay-fields and pastures, and near settlements. At lower elevations, up to approximately 1000 metres, there are mixed forests whose principal components are Lime Tree Tilia cordata and Beech Fagus silvatica, accompanied by various other broadleaved trees, by Silver Fir Abies alba and by Norway Spruce Picea abies; stands of Scots Pine Pinus silvestris are occasionally present in favourable places, especially on southern aspects. Higher up, spruce and fir generally dominate all the way to the tree-line, either alone, particularly at higher elevations and on northern exposures, or accompanied by beech. Larch Larix decidua is not particularly abundant, and occurs almost exclusively near the tree-line. Above the tree-line there are large expanses of alpine grassland, often used as summer pasture for cattle, interrupted by Rhododendron, Vaccinum and Alnus scrub, sometimes interspersed with Pinus mugo. The area sports a rich and varied herbaceous flora, with a high degree of endemism. This is particularly true for the limestone sections, but there are also notable acidophilic endemics, such as for instance Androsace brevis and Sanguisorba dodecandra.
2. Taxonomy. We generally adopt the names and scopes of taxa, as well as the subdivision in subgenera, given in Paul Williams' checklist of world bumblebees . In particular, we regard Psithyrus s.l. not as a separate genus, but as a subgenus of Bombus. However, we depart from the checklist in regarding Bombus cryptarum (Fabricius) as a valid species.
3. Species found. Altogether, I have observed 29 different species of bumblebees, a remarkable number given the small size of the territory explored. The table below gives, for each species found, the observed altitudinal range in metres above sea level and a rough (and mostly subjective) measure of frequency in the study area; four, three, two or a single asterisk mean, respectively, that the species is common, locally common or fairly common, uncommon, or rare.
Of the species found, Bombus barbutellus, B. gerstaeckeri, B. inexspectatus appear to be new for Lombardy.
4. Species accounts.
- Bombus barbutellus appears to be quite uncommon. I have found a female and several males in widely spaced areas of alpine grassland at fairly high elevation south of the Brembo-Adda divide, and several other males at lower elevations, in meadows surrounded by forest.
- B. bohemicus is exceedingly common throughout the area explored, from the lowest elevations to well above 2000 m, both in open areas such as pastures and alpine grassland and at the edge of forest.
- B. campestris seems to be local and fairly rare, and I have found only a limited number of males in two localities, two in a meadow at 830 m, and several others in another meadow at 1075 m. This is not so surprising, since its normal host B. pascuorum is also very uncommon in the area except at the lowest elevations. About half of the specimens found appeared more or less darkened, the others normal in colour.
- B. cryptarum is present with its subspecies reinigianus. It appears to be rare and has been found with certainty only in one locality at high elevation on the Sondrio side of the Brembo-Adda divide, foraging on Gentiana purpurea. However, I have seen males in other localities, on both sides of the divide but always at high elevation, which could have been this species.
- B. flavidus, which parasitizes species in the subgenus Pyrobombus, is rare; I have found just four individuals, all male, one on Silene acaulis at 2300 m on Cima Piazzotti, two others on thyme Thymus serpyllum at 2020 m near the Verrobbio pass, and a fourth one at 2200 m above the Verrobbio pass. All four have been observed rather early in the season, in late July and at the beginning of August. In Scandinavia, the primary host of B. flavidus is thought to be B. jonellus. In the Alps, where jonellus is rare, the main host is presumed to be B. monticola, with B. pyrenaeus suggested as a secondary host. The latter two species are both present where B. flavidus has been found.
- B. gerstaeckeri is uncommon but widespread. It occurs in much of the area explored, from as low as 900 m to well above the tree-line. Its presence is tied to the one of monkshood Aconitum, on which it forages almost exclusively. I have observed the species both on A. napellus and on A. lycoctonum, usually accompanied by B. hortorum or B. wurflenii. A. lycoctonum blooms earlier than A. napellus - up to a month earlier in the same locality - which must certainly be to the advantage of the bees, as it allows them to switch to a new food source when the first one is dwindling. It must also be observed that Aconitum often occurs in somewhat discontinuous linear colonies along streams, so that the bees can keep finding plants which are in full bloom just by moving upstream as the season progresses.
- B. hortorum, though not particularly numerous, occurs throughout, up to well above 2000 m, but is more common at lower elevations.
- B. humilis is fairly common to uncommon in meadows from 830 to about 1700 m, and occurs sporadically higher up. Most individuals belong to the all-dark subspecies apenninus, but I have also observed orange-backed individuals belonging to the subspecies propeaurantiacus, as well as a few individuals of the nominate subspecies.
- B. hypnorum appears to be fairly common, but localized, in and near forest at elevations up to 1600 m, and has been observed sporadically higher up. Together with pratorum, soroeensis, and, to a lesser extent, lucorum, it is one of the species which can be regularly found foraging on purple lettuce Prenanthes purpurea in forest.
- B. inexspectatus seems to be uncommon, but rather widely distributed in the area. I have positively identified one queen and several males. There have also been several sightings of individuals which can be tentatively attributed to this species, two of females, seemingly queens, and various others of males. One queen was found in rhododendron-alder scrub at about 1700 m, two others were observed in rhododendron scrub at approximately 1850 m. All the males were observed in alpine grassland at elevations between 1700 and slightly over 2000 m, except for three found at about 1500 m, one in a cleared area along a power line running through spruce-fir forest, two others in open areas along streams near forest. Individuals were observed both at the eastern and the western end of the territory explored, as well as in several places in between, but always on the southern (Bergamo) side of the Brembo-Adda divide. One gets the impression that the species is sparsely but continuously distributed throughout a rather narrow altitudinal belt between approximately 1700 and 2000 m, and occurs occasionally in open areas below tree-line. All the females observed were feeding on rhododendron, or not feeding, while most of the males were found on knapweed Centaurea nervosa and on various species of thistle (Carduus defloratus, Cirsium palustre, C. montanum). B. inexspectatus is almost certainly an obligate social parasite , probably of B. ruderarius, which is indeed rather frequent where I found inexspectatus. For the year 2005, when most of the observations took place, the pattern of these - few females, all seemingly queens, early in the season, and several males 30-50 days later - fits rather well with this hypothesis. It must also be observed that at the time when queens were observed - late June - colony development of ruderarius seemed to be well under way, as many workers and no queens were to be seen abroad.
- B. jonellus is rare, as I have found a single specimen in rhododendron scrub at 2000 m at the end of July. This particular individual was a worker of the subspecies martes.
- B. lapidarius is common at low elevations, below 1000-1300 m, and locally common in open sunny areas up to 1800-1900 m; I have found queens at over 2000 m, and workers foraging on thyme Thymus serpyllum at 2130 m on the southern slopes of Mt. Ponteranica, in an area normally frequented by montane species such as gerstaeckeri, monticola, pyrenaeus, sichelii and mendax.
- B. lucorum is very common everywhere, particularly below 1900-2000 m. It is frequent in light woodland and at the edge of woodland, but also in open habitats. It is often found on Rhinanthus, which it normally visits in association with wurflenii; it is also the most common visitor to heather Calluna vulgaris, at least near and above tree-line.
- B. mendax has been recorded, with a limited number of individuals, only in the most elevated parts of the study area around Mt. Ponteranica, where I have seen it foraging on various species of clover, on rhododendron, monkshood, and woundwort Stachys densiflora.
- B. mesomelas is rather common in grassland from 1250 m upwards, and is particularly frequent between 1600 and 2000 m.
- B. monticola is very common everywhere from approximately 1700 m upwards, and fairly common lower down. Together with sichelii and, to a lesser extent, soroeensis and wurflenii, it is one of the dominant species at high elevation, where it is by far the commonest species in ericaceous scrub.
- B. norvegicus seems to be rare; I have found just two individuals, both male, in a meadow surrounded by forest, near a stream, where its normal host, B. hypnorum, was also present, along with several other bumblebee species. However, since it is virtually impossible to distinguish the species from B. sylvestris in the field, it is hard to judge how frequent it might be without systematic random collections, which I have not performed.
- B. pascuorum is common only in the lower reaches of the study area, where it is ubiquitary. It becomes very uncommon above 1000-1200 m and occurs spottily up to slightly over 1500 m.
- B. pratorum is another common and ubiquitary species. It shares with hypnorum a fondness for woodlands, but is also regularly found in open areas, particularly at higher elevations.
- B. pyrenaeus has been found rarely, at high elevation and in rather cold, often north-facing localities. I have always observed it in humid places, mostly foraging on monkshood Aconitum napellus, but occasionally, particularly the males, also on composites such as golden-rod Solidago virgaurea.
- B. quadricolor is uncommon, despite the abundance of its host, B. soroeensis. Like this, it has been observed in widely different habitats, hay-fields at low elevation, pastures in a valley bottom, margins of forest and stream, alpine grassland, and alpine scree. Altogether, I have found five females and several males; however, the species can easily go unnoticed among the multitudes of red-tailed black bumblebees, and systematic random collections might reveal it to be more common than I have estimated.
- B. ruderarius is quite common in meadows and in alpine grassland above approximately 1300 m. It appears to be particularly frequent in a belt between 1700 and 2000 m. All the individuals observed belonged to the nominate subspecies.
- B. rupestris is quite common in meadows at lower elevations, much less so at middle elevations where forests dominate, and reappears in numbers above 1600 m, extending to the highest elevations. While all the females observed appeared to be quite constant in colours, males are variable, particularly in the amount of light hair on the thorax and on the first three tergites, ranging from rare individuals virtually without light hair, like females, to a form so light in colour as to superficially resemble, except for size, B. inexspectatus or a shaggy, long haired B. sylvarum.
- B. sichelii is one of the dominant species at high elevation, where it is more or less ubiquitary. It is widespread in open habitats from 1350 m upwards, but becomes really common, depending on local conditions, only above 1500-1800 m.
- B. soroeensis is probably the most common and widespread of the species observed, as it occurs in numbers at all elevations and in such diverse habitats as hay-fields, forest, alpine grassland, alpine scrub and scree. I have observed only individuals of the subspecies proteus; most of the females are all black except for a red tail, but one occasionally encounters individuals showing a rather narrow yellow collar, and sometimes also a yellow band, normally broken, on the second tergite. The species forages on a wide variety of flowers, but seems to be particularly fond of Phyteuma.
- B. sylvestris is widespread and quite common, though perhaps not as abundant as B. bohemicus. It can be found almost everywhere, from the lowest elevations to over 2000 m.
- B. terrestris has been found with certainty only below 1500 m, and appears to be uncommon. However, as in most cases I have not tried to identify to species individuals belonging to the subgenus Bombus s.s., it is quite possible that I have underestimated its presence.
- B. vestalis is rare, as I have observed just three specimens, all female. One has been found on the Sondrio side of the Verrobbio pass at an elevation of over 2000 m, in an area which, perhaps because of its very favourable exposure, is notable for the coexistence of species typical of higher elevations, such as B. mendax, B. sichelii, and B. flavidus, and species normally found lower down, like B. lapidarius. Two others have been found in a meadow at 1075 m.
- B. wurflenii is another very common species in the area explored. It occurs as low as 900 m, but becomes really numerous from 1200 m upwards, extending to 2200 m. It is found not only in open areas, but also in light woodland and forest margins. It is certainly the most common visitor to monkshood Aconitum at middle and higher elevations. It is also characteristically associated, usually in the company of B. lucorum, with Rhinanthus, which blooms profusely between 1500-1900 m on southern exposures.
5. Concluding remarks. A comparison between the list of the species found in the area studied here and the one compiled by Intoppa et al.  for the upper Val Camonica shows that the two regions have very similar bumblebee faunas, as was to be expected given the similarities between their respective climates and floras and the small distance (less than 60 km) and the absence of significant ecological barriers between them. The main difference seems to be the presence in the Val Brembana of inexspectatus and gerstaeckeri, which do not appear in the published Val Camonica list. However, it must be observed that the discovery of inexspectatus in the upper Brembo valley has prompted Intoppa and Piazza to re-evaluate some of their specimens from the vicinity of Edolo (Val Camonica), which they had initially identified as B. sylvarum; upon closer examination, these were indeed found to be inexspectatus . The discovery of inexspectatus in two localities of Lombardy fills a major gap in the distribution of the species in the Italian Alps. It can now be said with a certain degree of confidence that the species is probably present all along the Italian Alps at least from the Waldensian valleys of Piedmont to the Alps of Carnia in Friuli. Habitats like the ones where I have found the species abound all along the southern Alps and, given my experience in the area considered here, I see no reason to believe that the species could not be found in these as well. Incidentally, I have encountered inexspectatus also on the western slopes of the Corna Grande near Piani di Bobbio in the province of Lecco (also in Lombardy, and not very far from the localities considered here).
Also in the case of gerstaeckeri I am under the impression that the species might be more common and widespread in the Italian Alps than the scant literature records would lead one to believe. As I have said, the species is not too difficult to find and relatively numerous in the area considered in this note. I would find it a bit surprising if this were an isolated phenomenon, as I see no peculiarities of the upper Brembo valley that could justify the species being present there and not elsewhere. I might add that I have observed gerstaeckeri also on the Italian side of the Passo dello Spluga (Splügen Pass) in the province of Sondrio, on Aconitum napellus at 1900 m, and in the Val Ferret near Courmayeur in Valle d'Aosta, on Aconitum lycoctonum at 1700 m. On the other hand, there appear to be no literature records of gerstaeckeri either for Lombardy or Valle d'Aosta. On account of its peculiar diet, the species is very rarely seen before late July- beginning of August, and this might cause it to be under-reported.
Another apparent difference between the Val Brembana and Val Camonica lists is that the first includes three Psithyrus species, namely B. vestalis, barbutellus and quadricolor, which do not occur in the second. For what concerns vestalis and barbutellus this is not particularly significant, since these species seem to be fairly rare, and their absence from the Val Camonica list may not necessarily mean that they do not occur there, but just that they have escaped detection on account of their rarity. As for quadricolor, the species has actually been found in the Val Camonica  after the publication of  and .
The remaining differences between the Val Brembana and Val Camonica lists have to do not with the presence of species, but with their frequency. For instance, B. pyrenaeus is reported as common in the Val Camonica, but is rare in the Val Brembana, while the opposite is true for sichelii; probably as a consequence of this, B. rupestris, which parasitizes sichelii as well as lapidarius, appears to be much more common in the Val Brembana than in the Val Camonica. I do not know what this might mean.
A large portion of the Val Brembana bumblebee fauna consists of inquiline species - over one third of the total, if indeed inexspectatus is an inquiline. The valley hosts nine of the eleven European species of Psithyrus, and it is not unusual to find individuals of several of them foraging together; on one occasion I have found individuals of six different species in the same meadow, in close proximity of each other. Again, I do not know what the significance of all this might be.
1. Intoppa F., Piazza M.G., Ricciardelli D'Albore G. 1995. Catalogo bibliografico delle specie di Bombidae (Hymenoptera Apoidea) segnalate per l'Italia. Apicoltura 10, supplemento. pp. 135.
2. Intoppa F., Moreschi I., Piazza M.G., Bolchi Serini G. 1995. Bombus Latreille e Psithyrus Lepeletier del "Parco Naturale dell'Adamello" (Hymenoptera Apidae Bombinae). Boll. Zool. Agr. Bachic., Ser II, 31 (2): 167-178.
3. Williams P. H. 1998. An annotated checklist of bumble bees with an analysis of patterns of description (Hymenoptera: Apidae, Bombini). Bulletin of The Natural History Museum (Entomology) 67: 79-152.
4. Intoppa F. Personal communication.
5. Yarrow I. H. H. 1970. Is Bombus inexspectatus (Tkalcu) a workerless obligate parasite? (Hym. Apidae). Insectes Sociaux 17: 95-112.
6. Müller A. 2006. A scientific note on Bombus inexspectatus (Tkalcu, 1963): evidence for a social parasitic mode of life. Apidologie 37: 408-409.