N.Z. ECOLOGICAL SOCIETY 29 , thinner towards the north-east and on the The pH of the soil from Islington Bay is northern side of Rakino Island is about 7 6.9 in the topsoil and 7.3 in the subsoil. The in. thick-'--probably a thin dusting of it has base saturation of the topsoil is 83 per cent, influenced' soil fertility on the Noisies Is. and of the subsoil 6.3 per cent. Both of the further north. On both islands, it covers soils are youthful, having reached the nigre- fossil yellow-brown earths which,' on Motu- scent stage of soil formation, the main mor- tapu, appear' to have been podzolised in phological features being. a deep dark top- places, indicating that at some stage prior soil containing much well-decomposed to the eruption kauri had been a component humus and the absence of a B horizon, of the vegetation. The ash appears to be only slightly wea- A profile near Islington Bay is : thered, but the high base saturation figures Al () in. 'black sandy IOfLm with -tine gr'4llul>a,r structure, indicate that weathering is almost suffi- , ciently rapid to keep pace with the rate of .Raug;toto ash r Au 10 in. dark grey free medium , sand, leaching. Considering the age attributed to C 5 in. slightly comp&',t' dark grey- it, the soil appears to have undergone little 1 ish brown struC'ture.Jess sand, development. . 3 in. white slJ;ructurele...'*1 silt ] ] ' BurJ,ec poIlZO-" 100d loam, sOlI - { on yellowish cl'a.y. The lack of old forest roots in exposed profiles of the buried soils suggests the pos- On Rakino Island, the profile is: sibility of a scrub cover at the time of the Al, 7 in. black to dark grey-brown eruption and calls for further work. Rangit()
Forest Vegetation of the Inner' Islands of the 'Hauraki Gulf
I. A. E. Atkinson
Botany Division, D.S.I.R.
-. '; h }110RES'l' REMNAXTH A],\'D taraire-kauri forest pattern in which kauri , ( Agathis austra/is) was associated with the ., TH:": PRIMITIVE VEGETATIO:~.i , strongly leached and in parts podzolised ... .-' , northern yellow-brown earths of the upper The primitive vegetation of Auckland's larger inshore islal1ds, with the exception of valley walls and ridges; taraire (Beilsch- Rangitoto, has been either destroyed' or miedia tarairi) was associated with the greatly modified. Nevertheless, on some is- moderately leached northern yellow-brown lands, particularly Waiheke (26,000 acres) earths of, the sheltered coastal and inland and Ponui (4,400 acres), there are many valleys; and pohutukawa (Metrosideros ex- remnants of both primary and secondary celsa) occurred on the weakly leached and forest. When' considered, in relation to the skeletal northern yellow-brown earths of the present soil pattern and to early accounts, coastal slopes and cliffs exposed to 'wind- these indicate an originaV' pohutukawa- carried salt spray (cf. Taylor, 1960). 30 N.Z. ECOLOGICAL SOCIETY
are taraire, tawa, e.Motuhoropopo kohekohe, puriri, ma- .:POtato mangi (coprosma ar- ~ [) Moria I. borea), pigeonwood W Rokino , Horuhoru (Hedycarya aborea), '. (Gamet Rk,) rewarewa (Knightia MOTUi,\PU WAIHEKE exce/sa), and nikau (Rhopalostylis sapi- i: . .. ., .' "'~') .", daY. Houpara (Pseu- .'. Torokihi ,,::' dopanax lessonii) and ... :. ~hoql; .. .: ... kowhai (Sophora ~ '".. "" Pokot OQ RANGITOTO microphylla) are pro- ~ Retoroo minent near the coast. The presence' of secondary kauri is de- . . .0". '. pendent 0 n see d :( i sources and where, as on Ponui, large hol- - ' low trees were left
ary between the kauri and taraire com- was almost certainly not restricted to the munities.' coastline. Where direct archaeological evid- There are few forest remnants on Motu- ence is not available, some measure of the tapu (3,800 acres), Motuihe (450 acres), or extent to which land was cleared further Pakihi (270 acres). No sizable stands have inland may be possible from botanical evid- survived on Rakino (370 acres), Rotoroa ence. On many of the inland slopes and (220 acres), Motukorea (140 acres), Paka- ridges of Ponui and Waiheke, there are pre- toa (65 acres) or Karamuramu (12 acres). sent large-diameter pohutukawa trees which It is probable that the vegetation of these probably established before the arrival of islands was similar to that already described the European; annual ring counts are neces- though taraire would have been more pro- sary to verify this. Observations of relatively minent on the weakly leached red-brown undisturbed coastal forest on some of the loams of Motukorea. outer islands of the Hauraki Gulf have shown that pohutukawa does not normally On the smallest islands, pohutukawa, occur at any distance from the coast. It karo (Pittosporum crassifolium) and hou- does not establish under a closed forest para are the most important species. The canopy, or even in small canopy gaps caused vegetation of Tarakihi or Shag Rock, (12 by fallen trees. Thus it appears likely that acres), the Four Islands (David Rocks) and the trees mentioned above have established parts of Motuhoropapa (25 acres) and under high light .intensities with manuka Maria Island (4 acres), is apparently close (Leptospermum scoparium) and kanuka (L. ' to the primitive condition. Mr. P. A. S. Stein ericoides) following clearing of the original (pers. comm.) reports that vegetation on forest. the central part of Maria Island was de- stroyed by bombing practice during the war The few accounts of the early nineteenth but has since regenerated. The vegetation of century do not help greatly in obtaining a Otata (45 acres), the largest of the Noises clearer picture of Maori land use at this ~roup, includes mahoe (Melicytus rami- time. Marsden noted that part of Motukorea 1Iorus), mapou (Myrsine australis), manuka was under cultivation in 1820 (Elder, 1932), and flax (Phormium tenax) communities, and in 1827 D'Urville described the island most of which have developed following as "almost completely covered with a thick, fires. There are also a number of smaller carpet of very green grass". islets, some with unmodified vegetation, An intriguing question concerns the vege- where studies could be made to elucidate tation history ofMotutapu. The major event the relationships between the plants and in the island s history has been the eruption nesting sea-birds present. These include of Rangitoto in approximately 1200. + 50 Papakohatu (Crusoe Is.), Tombolo Is. A.D. (Fergusson and Rafter, 1959) which (Noises group), The Three Sisters, Horu- mantled both Motutapu and the neighbour- horu (Gannet Rock), and two unnamed ing Rakino with a layer of basal tic ash. This islets at the mouth of Putiki Bay, Waiheke. shower was still hot enough to carbonise , . vegetation when it reached the ground El<'I<'I'~C'l'S 01<' IV[AORI AND EmWPEAK SE'l"l'LKl\lK:":'l' (Brothers and Golson, 1959; Grant-Mackie, There is evidence of intensive Maori 1960), but so far carbonised remains of .settlementaround most of the coastline of large trees have not been found (cf. Taylor, 'these islands. Many terraced pa sites and 1960). This may be due to'insufficient search- 'ridges covered with pits, indicate the' extent ing but if not, then it is possible that even to which the coastal vegetation was altered. as early as 1200 A.D., Polynesians had made To the Maori, anembayed coastline would extensive clearings of the original forest. have, offered a sour~e of food from fishing, The presence of man on Motutapu at this headlands which, could be turned into de- time is shown by the finding of artefactual fensive positions;- and ,a means of communi- and food remains beneath the Rangitoto ash cation by.sea. He would,have required tim- by Mr. J. Golson (pers. comm.). There is as ber for stockade posts, c~floes, and firewood, yet no evidence to determine how large the apart from any land he Cleared for cultiva- population is likely to have been. . ' tl00. The arrival of the European in the early Alteraiion of the vegetation' by the Maori part of the nineteenth century initiated 32 N.Z. ECOLOGICAL SOCIETY widespread destruction of the remaining and Millener considered that 150-200 years large areas of forest. Cruise (1820) records could be taken as the maximum age of any visiting the inner islands in that year to tree on the island. Thus this evidence sug- collect kauri for spars. In the succeeding gests that plant colonisation of Rangitoto years large quantities of kauri were taken began somewhere between 1750 and 1800. from the eastern slopes of Waiheke and also from Ponui. Afterlogging came gum-digging, On the other hand, the exposure of Rangi- widespread fires and the establishment of toto ash described from Motutapu by pastures, which r~duced the forest to the Brothers and Golson (1959) shows only one steeper slopes an'd gullies. Manuka and primary ash shower, that of 1200 A.D., and kanuka communities established on sites this ash is not exposed above the lava flows where management was inadequate to main- of Rangitoto itself. On the evidence of tain pasture and Waiheke was for many human occupation levels in the Motutapu years one of the chief sources of Auckland's site, Brothers and Golson (loc. cit.) sug- firewood. gest that scoria cone building and lava emis- sion took place quite quickly after the ash Various mammals were introduced during eruption since it is unlikely that people the latter part of the nineteenth century. would have lived close to an active volcano. Rabbits and opossums established on most If this is the case there is apparently a period of the larger islands. Wallabies and fallow of 450-500 years following the eruption dur- deer were introduced to Motutapu and have ing which little or no plant colonisation took subsequently spread to the adjacent Rangi- place. toto. The large uncolonised areas of lava do 'rHF. PROBLEM 01<' RANGITOTO show that vegetation succession is extremely The vegetation of Rangitoto (5,800 acres) slow on some' parts of the gently sloping is a complete contrast to that of the other lava cone. These areas have remained bare islands and although in places it has been for at least 150 years, during which "time modified by fire or damaged by introduced pohutukawa seeds would have been blown mammals, the greater part is still little dis- on to the lava each year. The difference be- turbed. The island's communities constitute tween colonised and uncolonised areas can- the finest example of vegetation develop- not be explained satisfactorily by correla- ment on basaltic lava flows in New Zealand. tion with topography, since in many places These flows are of the aa type and are par- both lava ridges and gullies are covered with tially covered by a discontinuous short vegetation while in others they are bare. forest of pohutukawa (Metrosideros excelsa) and pohutukawa x rata hybrids (M. excelsa Part of the explanation for the discontinu- x M. robusta). The intervening areas of lava ous pattern of vegetation establishment may are often completely bare of plants, apart lie in the ocCurrence of localised ash erup- from lichens. tions subsequent to the lava outpouring. Mr. N. H. Taylor (pers. comm.) found what ap- The apparent youth of this vegetation was peared to be a young coarse ash soil over- pointed out by Millener (1953) who de- lying one of the outer scoria cones and it scribed a process of plant colonisation in may be found that the vegetation pattern is which isolated plants, usually pohutukawa, correlated with the distribution of ash establish directly on the broken surface of accumulations. However, this would not ex- the lava. Senecio kirk ii, puka (Griselinia plain the apparent long delay in initial col- lucida), fivefinger (Nothopanax arboreum), onisation. An answer to this problem may karamu (coprosma robusta), Astelia bank- have to be sought in the unfavourable but sii, and many other species establish in the variable edaphic conditions of the aa lava shade and litter of these initial colonisers flows themselves. Any factors which influ- thus forming "islands" of vegetation. Mil- ence moisture and available nitrogen levels lener (loc. cit.) estimated that the larger must be studied in relation to the establish- "islands" were spreading at rates averaging ment and growth of the vegetation "islands". 15 cms per year and that the rate of spread It will be of value to measure the number decreased with decreasing size of the "is- of new centres of plant colonisation which l~md". Ring counting gave figures up to 130, establish each year as well as their initial , N.Z. ECOLOGICAL SOCIETY 33 rate of growth. If a satisfactory explanation present on lowland areas of the Auckland is not possible along these lines, it may be isthmus. It is to be hoped that some definite necessary to, postulate a, more recent date programme of vegetation conservation can for the formation bf the lava flows. An exact . be implemented on these islands so that at geological dating for this event would avoid least representative stands are preserved for the dangers of arguing from botanical, evid- the benefit of future generations. ence alone. Dr. R. L. Bieleski (pers. comm.) has suggested that an endeavour be made ACKNO\\'LEDGKMI';KT~ . to collect carbonaceous material from be- neath the margin of the lava flows at their Stimulating discussions were had with contact with the old sea floor. Mr. N. H. Taylor of the Soil Bureau, D.S.I.R., and with Mr. J. Grant-Mackie, Dr. R. N. Brothers, Mr. J. Golson and Dr. L. H. Mil- CONCI..m.nON lener of the University of Auckland who, Apart from Rangitoto, there has as yet however, do not necessarily agree with ideas been little intensive work carried out on the expressed in this paper. I would also like vegetation of the inner islands. A number to thank Mr. P. B. Hannken and several of ecological topics worthy of further study other friends for contributing lists of plants emerge from the foregoing discussion. from localities on Waiheke which I had not visited. Miss C. T. Atkinson spent a consider- 1. AnalysiS and mapping of the larger able time in searching for historical accounts remnants of primary and secondary of these islands. Dr. R. L. Bieleski and Mr. forest with the objective of determin- N. T. Moar read the manuscript of this paper ing the factors responsible for develop- .and made many helpful criticisms and sug- ment of the basic pohutukawa-taraire- gestions. ' kauri vegetation' pattern and its vari- ations. .REFEIU;NGE8 2. Investigation of the history of this vege- Bml,J';SKI, R I.., 1959: }laewrs A free-Hng- Gl'Owth :uul tation with particular, respect to the Distribution of Rami (Ag(tthis aURtralis Halib.). 1. EffC(',t of Light on the E'~a.bli."Jll1neTht of Kauri amI length of time and areal extent of modi- of Phyllndadu8 trichomanoides D. Don. .-Iu,~t. .1. fication by Maori and Moa-hunter Bot. 7. settlement. There is an excellent oppor- BIWTIlERS, R. N.; GOT,HUN, J., 195!-): Goolol:,rical all(1 tunity here for ecologists and field arch- Arc.hacologiNlJ Interpret.ation of a Se<>tion in Ran{,.;- aeologists to carry' out co-operative toto .Ash on Motutapu Isla.ndJ Auckland. N,Z. J. studies. (}('ol. GcolJhys. 2: 5(;9-77. CnuJSr:, R. ..\., 19i37:"JQurnltl of (t Ten Jlo,nthlJ' ReNid- 3. Survey of the various introduced ani- (~nce in New Zea1and (1820)." The P{\gasu:; Pres;;, mals present on each of the islands and CJuistc.hul'ch. 2:W liP. f) 'U]WIJ.J,E, D., 1950: "New Zealand 1826-.1827." }~ng- determination of the extent to which }ish trans!. of the' Voyage de l' A~;tTolabe.' Wing- each is altering vegetation succession. fiold PI-eSH. 251 pp. ELlH;I{, J. n., 19;~2 (Edit.): "The Ldter.~ and ,JOU.l"nllls 4. Vegetation surveys of the smaller is- of Samu.d J/n/",Wen 176/1-1838." (:oull~ Somerville lands and islets with particular atten- Wilkie JAtl. '!mJ A. H. .Reod. 580 11p. tion given to the relationships between Flmm..:ssox, G. .T.; RA~-'I'lm, '1'. .A., 19,)9: Ntm' ZC~llan(! the plants present and colonies of nest- 14U Age Mcr.I."un~lI\eni:s--4. N.h. .T. Geol. G(;u}Jhys. ing sea birds. 2; 208-41. ~ , " Gn.\NT-:MACKIE, J. A., 1900: Outline Goology of t.he 5. Study of the edaphic conditions limit- Inner hhUltlS of the H~LUmkj Gulf. Proc. N.h. Reol. ~,. ing plant growth on Rangitoto, to- ooe. ""I: 2"d-.7 gether with a 'more detailed examina- KIRK, T., 1878: Notes on the Butany of Waiheke, ]~aJlgi- tion of the island's recent geological toto and Other Islands in the Haura.ki Gulf. Trans. N.Z. lust. 11: 44-4-;)4. history. ' M]Ll,~~.'\ER,L. H., :195::: How Old is t.he Vegetation on , " In the inner islands of the Hauraki Gulf, Rangitoto Island t N.Z. Eeol. Soc. Rep. 2nd Ann. Jleeting. biologists and naturalists have a readily TAYWR, N. H., 1960: Ob-serv1\..tions on the Soils of the accessible field for study which can throw Inner bhLlHls of tIle ll'UUI":.lkiGulf. ProG. N.Z. Ecol. light on the natural communities originally ~So.e. ' 7 .' _I-v.9"""