1 2 3 4 EPARATING LBATROSSES 5 S A 6 7 8 Tristan or Wandering? 9 10 Text and photographs by Peter Ryan 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 A Tristan fledgling on Inaccessible Island. Note the paler, greyer than 52 that of a of the same age. 53 54 55 Two types of Wandering Albatross occur off southern : the larger, 56 57 southern-breeding Diomedea [e.] exulans and the smaller D. [e.] dabbenena from 58 Tristan and Gough. They are now treated as separate by most 59 60 authorities, but how easy is it in fact to separate the two species in the field?  61 62 august/september 2000 tristan & wandering 35 1 1 2 2 3 3 4 4 5 5 6 A nine-year-old 6 7 (presumed female) breeding at 7 8 (far left) appears 8 9 similar to a dark, five-year-old 9 10 non-breeding Wandering 10 11 Albatross on Marion Island (cen- 11 12 tre), and is considerably darker 12 13 than another five-year-old non- 13 14 breeding female Wanderer (left). 14 15 15 16 16 17 17 18 18 19 19 20 20 21 21 22 sis. 8 500 pairs at Antipodes and Campbell islands, 22 23 lbatross has undergone some ; ranges through the South Pacific. 23 24 radical changes in the last five years, 24 25 expanding from 14 species to as many Gibson’s Albatross Diomedea [e.] gibsoni. rguments about the validity of these southern Africa. In the hand, the Tristan Albatross 25 26 as 24 species in four genera (see Africa – & 10 000 pairs at (south of New splits are ongoing, but this shouldn’t is smaller than a Wandering Albatross of the same 26 27 ABirding 2(1): 35). The great albatrosses have been Zealand); ranges through the South Pacific. deter birders from attempting to iden­ sex. But this is of no use in the field, and little has 27 28 most affected, with seven species in the complex: tify the great albatrosses as far as possible, if only been published on how to separate the various 28 29 Diomedea amsterdamensis. Ato provide additional information about the rang­ Wander­ing Albatrosses at sea. 29 30 Wandering Albatross Diomedea [exulans] exulans. 20 pairs at Amsterdam Island; range presumably es of these birds. During the summer of 1999/2000 I was fortu­ 30 31 14 000 pairs breed at South Georgia, Marion and South . Field criteria for separating the two Royal nate to spend several weeks on Gough Island and 31 32 Prince Edward, Crozet, Kerguelen and Macquarie Albatrosses are well known (summarised in Harrison three months on Inaccessible Island, which 32 33 islands; ranges throughout the . Diomedea [epomopho- 1983), and both species occur off southern Africa. together are home to all Tristan Albatrosses. Then, 33 34 This species is also sometimes referred to as the ra] epomophora. 13 000 pairs at Campbell and The is a rare but still in April 2000, I spent three weeks on Marion 34 35 Snowy Albatross D. [e.] chionoptera. Auckland islands and hybridises with Northern regular visitor to the continental shelf. The Southern Island, where there is a large population of nomi­ 35 36 Royals at Taiaroa Head, Otago, New Zealand; rang­ Royal is rarer, with fewer than five southern nate Wandering Albatrosses. Here are my prelimi­ 36 37 Tristan Albatross Diomedea [e.] dabbenena. es primarily through the South Pacific, but some African records, but recent observations suggest that nary conclusions about the possibility of separat­ 37 38 2 500 pairs at Gough and Inaccessible islands, and throughout the Southern Ocean. it is frequent in the Roaring Forties south of Africa. ing the two taxa in the field. 38 39 ; ranges through the South Atlan­ Recoveries of banded birds and measurements of 39 40 tic, and at least occasionally reaches . Northern Royal Albatross Diomedea [e.] dead birds confirm that both nominate Wandering 40 41 sanfordi. 8 600 pairs at Chatham Island and Taiaroa (exulans) and Tristan (dabbenena) albatrosses visit  41 42 Diomedea [e.] antipoden- Head; range as for Southern Royal Albatross. 42 43 43 44 44 45 46 47 48 A 17-year-old Tristan Albatross ä 49 (presumed male) breeding at Gough 50 Island (right) is slightly darker on the 51 wings than a seven-year-old non- 52 breeding male Wandering Albatross at 53 Marion (centre), and considerably dark- 54 er than a 13-year-old breeding male 55 Wanderer (far right). 56 57 58 59 60 61 62

36 tristan & wandering albatrosses africa – birds & birding august/september 2000 tristan & wandering albatrosses 37 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 of plumage pattern at sea is not helpful in sexing need to be examined. Wandering Albatrosses are 8 9 birds, apart from the very whitest males. Females characterised by considerable variation in plumage 9 10 are on average smaller than males, and with prac­ between individuals even of the same age and sex. 10 11 tice, given a good look at the bill, the sex can be In 2000/01 ornithologists will be based at Gough 11 12 determined fairly reliably. Males have noticeably Island for a full year, and one of their tasks will be 12 13 deeper bills than females, with less of a slender to gather more information on plumage develop­ 13 14 saddle behind the nail. ment in Tristan Albatrosses. 14 15 15 16 16 17 Tristan or Wandering? Implications for birders 17 18 18 19 Since the 1970s, several thousand albatross chicks Assuming that there is no easy way to separate 19 20 have been banded at both Gough and Marion Tristan and Wandering albatrosses in the field, 20 21 islands. I compared known-age and -sex birds and birders might be tempted to reject the notion of 21 22 described extreme forms in the breeding popula­ recognising them as distinct species. But this has 22 23 tions. The photographs on pages 36 and 37 tell the some interesting consequences. The genetic evi­ 23 24 story. In brief, faced with a single at sea, of dence suggests that the Amsterdam Albatross is 24 25 unknown age and sex, I could find no definitive most similar to the widespread, nominate 25 26 character that allows separation, but there are Wandering Albatross (Nunn & Stanley, 1998). The 26 27 some possible differences that require further two New Zealand forms, gibsoni and antipodensis, 27 28 investigation. form another species pair, with the Tristan 28 29 Juvenile Tristan Albatrosses leave the nest in a Albatross being the most distinct form. The point 29 30 paler, greyer plumage than Wandering Albatrosses, is that if we recognise the Amsterdam Albatross as 30 31 but we don’t know how this plumage changes a ‘good’ species – which most birders are happy to 31 32 until the birds return to their breeding islands. do – then we should recognise at least the Tristan 32

33 © peter ryan Tristan Albatrosses take longer to replace the juve­ and New Zealand populations as distinct from 33 34 nile plumage than Wandering Albatrosses. Most southern, nominate Wanderers. 34 35 breeding Tristans have a large amount of black on Unfortunately, this type of problem is going to 35 36 the upperwing (corresponding to Harrison’s stages become ever more prevalent as types of evidence 36 37 4 and 5), whereas most breeding Wanderers at not readily apparent in the field (for example 37 38 Inferred plumage sequence of the Tristan Albatross, showing upperparts (above) and underparts (below), Marion are stages 5 to 7. However, I did see a few genetic and skeletal characters) are used to define 38 39 based on field sketches and photographs. Birds become progressively paler from fledglings (left) to old Tristans on Gough that had almost all-white wing bird species. There is no easy solution for birders – 39 40 males (right). Note the retention of prominent breast bands on birds with pale backs. coverts and no dark feathers visible in the tail the best we can do is to use our birding skills to 40 41 (stage 7, and see figure on page 38). The same pat­ try to detect visible differences, and learn to 41 42 tern of delayed whitening occurs among the two accept that it may not be possible to identify 42 43 New Zealand forms, Gibson’s and Antipodean alba­ some birds to species.  43 44 trosses. 44 45 Tristan Albatrosses begin breeding at an earlier 45 46 age, starting as young as six, with the average age Further reading 46 47 Complex plumage patterns On the underparts the pencilling often forms a of first breeding being eight to nine years, com­ Harrison, P. 1983. : An Identification 47 48 breast band, with some brown mottling persisting pared with 10 to 12 in Wandering Albatrosses. But Guide. Croom Helm, Beckenham, Kent. 48 49 The Wandering Albatross is well known for its on the thighs and undertail coverts. Pencilling also this doesn’t account for the darker plumage of 49 Nunn, G.B. & Stanley, S.E. 1998. ‘Body size 50 complex series of age-related plumage changes. replaces the mottling on the back and rump, adults. The nine-year-old presumed female photo­ 50 effects and rates of cytochrome b evolution 51 Unlike most birds that acquire a fixed adult plum­ although a dark cap (or vestiges thereof) may be graphed on Gough is at least as dark as the darkest 51 in tube-nosed seabirds.’ Mol. Biol. Evol. 15: 52 age, the Wandering Albatross continues to change, retained into early adulthood. four- to six-year-old pre-breeding birds at Marion 52 1360–1371. 53 becoming progressively whiter until it is 20 to 30 Once the back is only lightly pencilled, the Island. Similarly, the 17-year-old presumed male is 53 54 years old. Fledglings are brown all over except for upperwings begin to whiten from the elbow joint, similar to a seven-year-old Wandering Albatross. Prince, P.A., Weimerskirch, H., Huin, N. & 54 55 the underwings, face and throat. With each succes­ in the centre of the wing close to the body. This Unfortunately, this difference is of little use if you Rodwell, S. 1997. ‘Molt, maturation of plumage 55 56 sive moult, the brown body plumage is replaced forms a ragged white oval which gradually spreads don’t know the age and sex of the bird you’re and ageing in the Wandering Albatross.’ 56 57 with white feathers. This happens rapidly on the until all the coverts are white. With time, the pen­ attempting to identify. 99: 58–72. 57 58 underparts, but is more protracted on the upper­ cilling on the body plumage gradually fades, leav­ Among adult albatrosses, the only possible fea­ Weimerskirch, H., Lequette, B. & Jouventin, 58 59 parts, giving a mottled appearance (the so-called ing the oldest birds all white except for the dark ture I found was that Tristan Albatrosses retain the P. 1989. ‘Development and maturation of plum- 59 60 ‘leopard’ stage). flight feathers. Males whiten much more rapidly pencilling on the underparts longer, relative to the age in the Wandering Albatross, Diomedea exu- 60 61 The body plumage then becomes almost all than females, and attain a white final plumage, but upperparts, resulting in a prominent breast lans.’ J. Zool., Lond. 219: 411–421. 61 62 white with fine dark barring, termed ‘pencilling’. old females are whiter than newly adult males. Use band. However, more birds of known age and sex 62

38 tristan & wandering albatrosses africa – birds & birding august/september 2000 tristan & wandering albatrosses 39