by Dave Goulson
After Malargüe, the route left the vineyards behind and wound through rocky desert for nearly 160 kilometres before coming to the crossing of the Rio Grande. The broad, fast-flowing brown river emerged from a steep-sided rocky ravine to the west, spilling into a flat-bottomed valley where it braided into numerous channels between banks of gravel. On the far side of the bridge lay a dead horse with what appeared to be a fence post protruding from its abdomen. Just beyond the ill-fated beast was a small cluster of half a dozen simple houses amongst a few acres of verdant green fields which stood out in the otherwise arid brown landscape. It appeared to be a likely spot for bees, so we stopped to search. The fields were well defended behind formidable fences made by weaving together the thorny branches of acacias – perhaps to keep pumas from the livestock within. They were also effective at excluding entomologists, so I skirted around the little patch of green fields along the shingle of the river bed. Peeking through the fence at a little patch of overgrown pasture, I noticed flowers of a distinctive bluish colouration – it was unmistakeably viper’s bugloss. If there were any bumblebees here they would surely be on this, as it is a flower much-beloved by bees of all sorts for it produces copious quantities of sugar-rich nectar. In my mind, viper’s bugloss is intimately associated with happy summer days hunting for bees, for it is found in most of the very best places for bees in Britain: Salisbury Plain, the shingle and dunes of Dungeness and the coastal dunes and marshes of the Thames Estuary. It is an invasive weed in New Zealand, in the drier parts of which it has taken over large areas of sheep pasture, covering the landscape in a blanket of cobalt-blue flowers and forming a mainstay of the diet of the British bumblebees that now live there.fn5 I always have a patch in my garden, for it is easy to grow and looks magnificent in the herbaceous border. It is even said to be good for treating snake bites (presumably the origin of the name) and bee stings, though I have never tested this.
Luckily, there was a point where I could clamber over the thorny fence, so I did so a little sheepishly, expecting an irate farmer to shout at me at any moment. There were swarms of bees – carpenter bees, honeybees, all manner of solitary bees, and bumblebees. A moment’s inspection revealed that these were the buff-tail invaders – not just one or two, but dozens of them, everywhere, feasting on the viper’s bugloss. We had met their vanguard, advancing north from San Martin.
In some ways it was lovely to see such familiar creatures. Buff-tails are by far the easiest bumblebee to breed in captivity, so I have spent many happy hours studying them in various experiments over the years. But although we knew we must meet them somewhere along our route, I had hoped that it would not be quite so soon. A quick look on the map confirmed that we were still about 900 kilometres north from San Martin, a distance which the buff-tails had covered in just eight years. Given that they normally have only one generation each year, and that the nests themselves clearly cannot move (unless they are cleverer than I suppose), all of this ground must have been covered by queens before they built their nest each year, each dispersing on average just over 100 kilometres. This is extraordinary – in Tasmania and New Zealand, the buff-tails spread much less rapidly following their introduction, covering just a few kilometres each year, although they did eventually colonise the entirety of both islands. We had had no idea that queen bumblebees were even capable of such epic flights. Although these were an alien species, potentially doing terrible harm to the indigenous bumblebees, I couldn’t help but feel affection and not a little admiration for these hardy little creatures, such a familiar sight in Europe, forging a successful life for themselves in this new world. It was, after all, not their idea to come to South America in the first place, and we can hardly blame them for making a success of it.
I called Jess over, and between us we caught and pickled a sample of bees so that we would be able to screen them and determine which diseases they were carrying. The bumblebees were so numerous that at one point I accidentally caught two in my net at once, a schoolboy error. Getting a bee from a net to a pot is fairly straightforward with a little practice, but fraught with difficulty if there is a second angry bee in the net at the same time. A sensible person lets them both go, but I can never resist the challenge of trying to get them both into a pot simultaneously. The outcome is almost inevitably a stung hand, and so it was on this occasion. Fortunately, Jess proved to be particularly adept with a net and caught plenty while I was floundering around, cursing and sucking my throbbing thumb.
We set off southwards once more, following the Rio Grande for many miles. Despite the water there were few flowers and no bees. We entered a spectacular area of active volcanoes, with several neat, conical peaks a little to our west. The river had carved its way through jet-black, solidified lava flows, and our road deteriorated to a rough gravel track as it attempted to follow the river through this magnificent but inhospitable landscape. It was baking hot and the black rocks soaked up the sun’s heat, creating a shimmering heat haze. There was almost no plant life, indeed almost no life of any kind apart from house flies, which descended on us in droves the moment we stopped to eat our picnic lunch. I had had a similar experience when sitting in the middle of the Sahara thirty years earlier. Quite how these creatures manage to sustain themselves in such places is utterly beyond me.
After 250 kilometres or so we emerged into a welcome patch of greenery surrounding the small town of Chos Malal, and here we immediately found buff-tailed bumblebees in abundance. I couldn’t help but wonder how the buff-tails had managed to cross the landscape we had just come across. It had been mildly arduous for us, with the benefit of an air-conditioned car, lots of bottled water and a nice packed lunch – how a queen bee could find her way across that flowerless wasteland was hard to conceive.
From here on, southwards to San Martin, the story remained much the same. The road travelled through several more expanses of rocky, arid desert, but wherever there was a patch of greenery there were numerous buff-tailed bumblebees, and no matter how hard we searched there were no native bumblebees to be found, and in particular, there were no giant golden ones. We would catch and pickle a sample of buff-tails, then move on. As we drove we pondered why the buff-tails had become so common here. They presumably benefited from the plentiful European weeds, such as the viper’s bugloss and thistles, that we saw growing along the roadsides. Perhaps they also enjoyed freedom from competition with the many other short-tongued bumblebee species found in their native European range, such as in Poland. It seemed likely that they would have left behind many of their natural parasites, creatures such as the tracheal mites that infest their airways and conopid flies that eat them alive from the inside. Of course once we were back in the UK our samples of bees would allow us to find out exactly what parasites and diseases they did have.
Some eighty kilometres before San Martin the horizon ahead of us acquired a murky, grey appearance. It looked a little like the dense smog I have seen above cities such as Los Angeles and Seoul. The road, which had been taking us across a rocky plateau, suddenly plunged over the lip of a mighty ravine, and zigzagged perilously down the cliff face on narrow ledges. I pulled over. It should have been a most splendid view, a slightly less grand version of the Grand Canyon, except that the air was thick with what I took to be smoke, and we could only occasionally see across the valley as the smoke ebbed and gusted in the brisk wind. We got out, and it was immediately clear that it was not smoke but grit and dust, which stung our eyes and made us both sneeze. Huge black birds soared on the updraft, eight of them in total. It was hard to see them well, squinting into the gritty wind, but I noticed that they had white ruffs around their necks and that they were further away than I had at first realised, and hence they were larger – they were in fact condors, the largest flying birds on earth. It was a spectacular and eerie sight, these monstrously large vultures hanging silently in the swirling dusty air.
We watched them for as long as our eyes could bear, and then headed down to the canyon floor. Soon after
wards we passed through the small town of San Junin and the vegetation changed dramatically. Lush green forests appeared in the valley bottom, the first forests we had seen in a journey of roughly 2,500 kilometres, and as the road wound westwards we emerged into a rolling landscape of forested hills, clear babbling streams and icy blue lakes. Were it not for the dust which still hung in the air, it would have been exceptionally beautiful. These forests stretch south to San Martin and San Carlos de Bariloche beyond. Most of Argentina lies in the rain shadow of the Andes – the prevailing winds come from the west and drop their rain on Chile as they rise over the Andes, so that the air is dry by the time it sweeps down the eastern side of the mountains. However, in this area the mountain chain is lower, allowing moist air from the Pacific to reach this far east and support these lovely forests. This, of course, is also why this is the area where introduced bumblebees, both ruderal and buff-tails, first made the crossing from Chile to Argentina.
We stopped once more to search for bees. There were plentiful flowers, and buff-tails were abundant. Spectacular swathes of orange-flowered alstroemeria and purple fuchsia grew among the trees. I caught the first and only ruderal bumblebee that we saw on the entire trip, a queen feeding on a dusty hollyhock flower by the roadside. Until 2006 the giant dahlbomii had been a common sight in this area but, as we had been led to believe, there were none to be found.
As we travelled south-west the dust became thicker, and the bees less frequent. The ground became covered in a layer of it, the vegetation smothered under a blanket of pale grey, and it finally dawned on me that this was ash. We had heard that a volcano had erupted just over the border in Chile some two months earlier, and that it was continuing to throw out pumice and ash, but coming from a land where such things are unknown, it had taken me a while to recognise what I was looking at. We passed lakes which were partially covered in layers of pumice at the leeward end – pumice floats, and we later learned that, just after the first and strongest eruption, it was possible to walk across these lakes, so thick was the floating layer. The few buff-tailed bees we saw looked pretty sad; they were themselves covered in dust, and were trying to forage on ash-covered flowers. Their pollen loads must have been mixed with ash, and I suspected that this may have killed the developing grubs, or else perhaps many of the nests were simply buried under ash and the bees unable to escape. Certainly bees of any type were few and far between.
We passed through San Martin, which proved to be a pretty tourist town on the edge of a long lake, popular for hiking in summer and skiing in winter. The houses had mostly been built by Swiss and German settlers out of timber and styled upon Alpine chalets, making the area seem like a slice of Switzerland plonked down in South America, an impression heightened by the surrounding snow-capped mountains. We stopped for a drink in a café that sold nothing but twenty-five variants of hot chocolate, plus huge slices of delicious chocolate cake to accompany it.
In Bariloche, a day’s drive beyond San Martin through beautiful forested scenery during which we left most of the ash cloud behind, we met up with Carolina and her PhD student Marina Arbetman. Marina had recently started her PhD on the impacts of the invading buff-tailed bumblebees. She had been trying to chart their spread southwards from San Martin, and had recorded them as far away as El Calafate, some 1,200 kilometres to the south. She had seen almost no dahlbomii anywhere within the expanding range of the buff-tails, but a colleague had reported seeing one a few days earlier by a nearby lake. Marina took us to look for them. There was much less ash here and the area was stunning, with jagged-toothed snow-capped mountains framing the glorious turquoise lakes. The area is protected as a national park, swathed in dense forests, and rich in wildflowers. We walked along the shore of Lago Nahuel Huapi, at the place where the dahlbomii was recently seen and where Carolina has been regularly counting bumblebees for many years. The shore was lined with Arrayan trees (Luma apiculata), gnarled and twisted by the fierce winds which must blow across the lake in winter. The trees were covered in delicate white blossom, flowers formerly favoured by dahlbomii, but all we saw were buff-tails. There were plenty of other things to see. We spotted pigeon-sized blue, white and rust-coloured ringed kingfishers hunting by the lake shore, and flocks of noisy olive-green parakeets flew past. Nonetheless, it was hard not to be a little depressed. From Mendoza southwards for 1,300 kilometres, we had been within the historical range of dahlbomii, but we had not seen a single one. Now it was clear that they were close to extinction throughout almost all of their known range. Beyond El Calafate, it is just another 300 kilometres to Tierra del Fuego, and then about the same distance again to the very tip of South America. At the current rate of spread, it seemed likely that buff-tails would reach the tip of South America within two to three years, and that might be the end for dahlbomii.
So what is causing the decline of the giant golden bumblebee? We discussed this at length with Marina and Carolina. It seems unlikely to be competition from buff-tails. The difference in their tongue lengths means that they have different flower preferences, although they do feed on some of the same flower species. Long-tongued and short-tongued species readily coexist elsewhere in the world, and in places such as Poland dozens of species happily live alongside one another. Also, the speed of decline of dahlbomii seemed to be far too fast to be explained by competition.
We discussed the possibility that it might be due to disease. Little is known about bumblebee diseases in most parts of the world, but we do know that they can suffer from a diverse range of infections caused variously by protozoans, bacteria, fungi and viruses. When the Spanish conquered the new world they accidentally infected the native Americans with a host of European diseases to which they had little or no resistance. Diseases such as measles, from which Europeans usually recover swiftly and without any permanent ill-effects, proved to be lethal. The impact was devastating, with perhaps as much as 95 per cent of the entire population of the Americas wiped out over a few decades. Entire civilisations, such as that of the Aztecs, collapsed into chaos and then were easily overrun by a tiny force of European invaders. Recent evidence even suggests that substantial parts of the Amazon basin, long thought to be primary, undisturbed rainforest, actually supported large civilisations and extensive areas of farmland until about 500 years ago when they were wiped out by these diseases, which spread far ahead of the Europeans themselves. The process was largely one way, but the American natives did give the Europeans syphilis in small revenge for their fate.fn6
Was something similar going on in bumblebees? My PhD student Pete Graystock had recently performed some ingenious experiments in which he had shown that bee diseases can be transmitted between bees of different species without the bees actually ever needing to meet. Every time an infected bee visits a flower she contaminates it with a sprinkling of parasites, perhaps contaminating the nectar with her mouthparts, or the petals by contact with her feet and body. The next bee that comes along may unknowingly pick them up, and either ingest them or carry them back to her colony. Pete came up with the neat term ‘florally transmitted diseases’, FTDs for short. If buff-tails were carrying something to which they were largely resistant, it could easily spread to native bees. However, testing this hypothesis is difficult. There are virtually no dahlbomii left, so we can’t study them to see what is killing them. If one were lucky enough to catch one of the few survivors, it would presumably be one of the few that had escaped the disease, or perhaps one of a small number with natural resistance. Studying the invaders can tell us what diseases they have – this was why we had been catching samples of buff-tails – but it cannot tell us which disease is killing the natives, if indeed it is a disease.
Marina had already started work on this, and had found that many buff-tails near Bariloche were infected with Apicystis bombi, a disease known from bumblebees in Europe and North America. In Europe this disease is sometimes fatal to bumblebees, but many individuals seem to carry the pathogen without showing symptoms. This disease had not been
recorded previously in South America, but on the other hand hardly anybody had ever looked so it is hard to know what this means. It is quite plausible that this disease arrived with the buff-tailed bumblebees and is the cause of the dahlbomii decline, but we cannot say for sure.
We subsequently studied the pickled buff-tails that we collected in Argentina and found many of them also to be infected with a second disease, Crithidia bombi. This is a trypanosome, a relative of the sleeping sickness spread by tsetse flies in Africa, which causes gut infection in bumblebees. Infected bees suffer a range of symptoms – their ovaries tend to be smaller, and they seem to be a little more stupid, being less well able to learn and remember which flowers are most rewarding. Crithidia bombi had been found in South American bees before buff-tails were introduced, but the invaders were carrying genetically distinct strains that presumably came with them from Europe. Maybe this is the cause of the collapse of dahlbomii?
Clearly more work is needed to find out exactly what is going on. It should be possible to use museum specimens of dahlbomii to see what diseases they used to suffer from before buff-tails arrived. It might sound tricky to tell what diseases a long-dead bumblebee on a pin might have been carrying, but modern genetic techniques are very sensitive and ought to be able to detect the DNA of the pathogen. Marina has already had a go at this, and she has not detected any Apicystis in a small sample of old dahlbomii, but she did not have many specimens to play with and she could not be sure that the negative result was genuine or whether her technique was faulty. What she really needed were old museum specimens of some bumblebees which we knew had been suffering from Apicystis when they were killed as ‘positive controls’, but no such bees are available.
