Liriodendron chinense and tulipifera

I saw a young Chinese Tulip Tree, Liriodendron chinense (Hemsley) Sargent,  on the KMOTT visit to the Hill Farm Oast Arboretum garden on July the 7th, a fairly young tree but well established as most trees there are.

The flowers are similar to the American species, Liriodendron tulipifera (L.), if perhaps rather less orange and colourful, but the leaves of the tree are more attractive, with clear waists between two neat pairs of lobes, as also often seen in coppice shoots of tulipifera.  The RHS describes it as potentially bushy, with flowers only found on mature trees, and they claim usually propagated by grafting, which is a surprise. It may not be quite as hardy as the American species, and tends to be grown only occasionally in the UK, although planting is now increasing. It is said to be a bit smaller than the American tulipifera, which grows very large, one of the tallest trees in American woodlands. Its native range is mid-China, extending down to Vietnam, but it is decreasing rapidly due to logging and habitat loss.

Liriodendron chinense has been awarded the RHS's Award of Garden Merit, AGM. The leaves in this species flush a bit red-purplish unlike tulipifera, and there is always good reliable golden-yellow autumn colour, as in tulipifera. There is also a cultivar called J.C. Raulston, from North Carolina, with larger and darker leaves.

It wasn't seen by a westerner until 1875 and was then introduced to the west by 'Chinese' Wilson in 1901, at the turn of the 20th Century, and there are actually no known further introductions of seed until 1977 when seed was gifted from Ninjing Botanic Gardens to Kew, although several seed introductions since then have allowed it to be more commercially available and more widely planted. Seed can even be found on Amazon and eBay.

There are now said to be only two remaining (extant) sister species of the old genus Liriodendron, which are said to have diverged in the late Miocene period. Liriodendron is a genus in the primitive Magnoliid group within the Angiosperms, which is now thought to be quite ancient and separated from the much larger groups of the Eudicots and the Monocots.

Genetic diversity analyses of the DNA suggest that L. chinense has tenfold higher genetic diversity than L. tulipifera, suggesting that the complicated regions comprising east–west-orientated mountains and the Yangtze river basin (especially near 30° N latitude) in East Asia offered more successful refugia than the south–north-orientated mountain valleys in eastern North America during the Quaternary glacial period.

The same paper also suggests that L. tulipifera lies in between the Eastern and Western clades of L. chinense. I cannot quite see how that makes sense in terms of two divergent sister species. There are  also interspecific hybrids as one might expect.

The Chinese Cork Oak, Quercus variabilis, Bedgebury.

There are some nice young trees scattered about Bedgebury, and I rather think this is one of them, above one of the paths going away from the visitor centre.

The Chinese Cork Oak, Quercus variabilis Blume, is found across the Far East - China, Taiwan, Korea and Japan, and has sometimes been used commercially for cork production - although the productivity and quality is rather poor. In Korea there are trees that can be seen with evidence of cork removal from the trunks up to "ladder-height". It is rarely planted in the UK and USA, but it may be considered as fairly ornamental, and young trees hold their leaves into the start of the winter period. It was first introduced to England by Robert Fortune in 1861.

The leaves are quite Sweet Chestnut-like, ovate-lanceolate, but very bright green and particularly glossy, with the obvious veins (13 - 18 pairs) ending in bristle-like extensions. The petioles are relatively long, often several inches. The Collins book hasn't quite got the margins of the leaf right, making the lobes too "rounded". However the specific epithet - variabilis - does refer to variability in the leaves, so perhaps I shouldn't be over-critical. The twig has been said to be somewhat shiny, as it appears here.

Importantly the underside of the leaves are grey-woolly overall, not just on the underside of the veins, but I don't have a good picture of that. But, here are some new, and even replacement, leaves developing.

The bark on the trunk is corky from an early age - these trees are 16 years old (on this site) and are also bearing acorns (which I would love to collect this autumn - if I remember!) The acorns are very rounded or semi-globose and fairly fully covered in curved mossy bristles.

The winter buds are said to be pointed. The tree is thought to be hardy, and to tolerate some shade. It is said to prefer moist loamy or clay soils.

The Indian Horse-Chestnut, Aesculus indica

the Indian Horse Chestnut, Aesculus indica

The Yellow or Sweet Buckeye, Aesculus flava

This lovely tree's natural range is the Ohio Valley and the Appalachian Mountains, USA. It was introduced to the UK in 1764. The tree I have been looking at is on the front lawn of Hadlow College main campus, which is apparently not grafted. I also had a quick look at the grafted and really quite large tree at the Gravesend Cemetery.

The flowers on this tree are a pale creamy yellow, with yellow flashes inside the two upper petals which soon turn pinky-red as the flowers age. As the flowers age further salmon-pinky tinges appear on the outside of the petal tube. However, I do not know if the flowers should not have any red in them at all.

Importantly the stamens are not exerted outside the petal tube, which character distinguishes this species of tree from the Ohio Buckeye, Aesculus glabra, one of the other Aesculus species with largely yellow flowers. Eventually however the style is exserted. On very limited evidence I would say that Aesculus flava flowers with Aesculus indica one or two weeks after Aesculus hippocastanum and Aesculus glabra, but I cannot be sure of this as the Hadlow tree was flowering at the start of June in 2020, while the tree in Gravesend was over flowering and only showing developing fruit. Maybe the leaves are slightly up-folded - cupped in this photo of the Hadlow tree??

These flowers are ageing a bit and their flashes have already turned red. The outer part of the petal tube is beginning to turn a bit salmon-pinky as well.

The leaves are very neat, appearing glaucous (blue-green) and woolly overall underneath and also specifically hairy on the top of the midrib of the leaf, although the top of the leaf is a good mid-green, contrasting quite strongly with the underside. Some authors say that the pubescence of the underside is quite variable. There are five leaflets generally, not seven, at least as far as I could tell on this tree. This seems quite a controversial point, Mitchell sees it as partly diagnostic just having five leaflets, while others think this can vary up to 7. However Mitchell also says that the leaflets are fully stalked, noted to be glossy, and often cupped, none of which I can really fully see in this tree. I don't think the leaves are fully glossy but the stalked leaflets is the thing I most take issue with, as the leaves is the Sibley book look sub-sessile, like the Hadlow tree. Maybe it would be good to walk the hills of Ohio and see the variability of the trees that actually exist in the full population. It would also be interesting to see the "hybrid zone" containing Aesculus x marylandica on the boundary between Aesculus flava and Aesculus glabra. Quite a task!

The feature of "leaf cupping" that is mentioned by Mitchell and on line might be this upward folding of the leaves that I think you can see in this photo of the Hadlow tree.

Looking at the Gravesend tree, which I didn't photograph, it fits the "standard" description much better - the leaves a bit more cupped, they are a little glossier, and the leaflets definitely have noticeable stalklets or petiolules. They are a bit more broadly obovate, while the Gravesend tree has leaves a little bit slimmer and more elegantly pointed at the tip. Could it be that the Hadlow tree is somehow atypical, possibly even containing a small amount of Aesculus hippocastanum?

The large lumpy spineless fruit should normally contain 1-3 seeds (conkers).

The leaves are said by Mitchell to give good bright orange-red autumn colour, Collins and More have it as even redder, being the reason why it is planted, but Braakman at the National Arboretum in Holland says that there is no autumn red colouring in the pure species. All very strange!

The twigs are said to have a bit of an odour, but not the rank offensive smell of Aesculus glabra. The branches are often said to be twisting and ascending.

The bark of the species is said to be a smooth dark brown (Bean's) or grey-brown with some scaly flaking. The trunk of this Hadlow tree has been damaged to the southwest, with flaking bark, but I am not at all sure by what. So I must travel to see other trees planted within striking range.

According to Trees and Shrubs online, there is a form called A. flava f. virginica, in which the flowers are red instead of yellow. This form was supposedly collected by George Washington for his garden in Mount Vernon. This source views such trees, "if in cultivation", as possibly distinguished from A. × hybrida, the reddish-flowered cross between Aesculus flava and Aesculus pavia by the absence of glands from the edges of the petals. However the Holden Arboretum
https://holdenarb.org/horticulture/plant-profiles/aesculus-x-hybridia/ views these two taxa as synonyms for the hybrid - in other words all these trees are the hybrid, with presumably the glands present on the sides of the petals. The hybrid has exserted stamens.

Another possible variety is f. vestita, which is pubescent under the leaves and along the young branchlets. Found in various localities in the wild (Bean's)

It is said to prefer acid to neutral soil. In the UK it is suggested that shelter may help to get it established and that it does better long term when the top of the canopy can be in full sunshine.

The Rauli, Nothofagus alpina

I found two lovely young "Rauli" trees this morning at The Grove, a woodland from the old Redleaf Estate. The Rauli or Alpine Southern Beech is known scientifically as Nothofagus or since 2013 Lophozonia alpina (or nervosa or procera) 

The apparently deciduous leaves were beautifully oval with rather variable tiny serrations set over an undulate margin. The key from Gandolfo and Romero, 1992 mentions that the sinus of the leaf margin lies just beneath the vein end rather than just above the vein end. They were mid-green on the upper surface, paler beneath. The upper surface has tiny "spots" on them, looking rather like miniature rivet heads, and these may also represent the tiny hollows apparent on the underside of the leaves. It seems strange that these are not commonly mentioned in the texts.

The leaf veins are quite deeply impressed, and 15 - 20 pairs seems quite common. The shape of the leaves and the numbers and appearance of the veins together are quite diagnostic.

The (initially green for a year) young shoots and also the petioles, up to 12 mm long, are covered in tiny brownish felt-like hairs. Some of the upper surface, such as the midrib, the margins of the leaves and the underside of the leaf veins are also covered in small silvery hairs.

The buds at this time of year are quite small, a mm or two long, but should be much larger by the wintertime, maybe a cm or so in size.

The tree comes from moderate altitude in the Andes mountain range in Chile, extending into the coastal ranges, and also with a few stands just over the border into Argentina. It is quite far to the south, 35 to 42 degrees. It was introduced to the UK between 1910 and 1914 by F.R.S. Balfour of the Dawyck Estate. In this country it appears moderately hardy, with some losses of younger trees in hard winters. Once over 20 years old you can get fairly good seed crops. It generally does better in the wetter west of the UK.

The tree has a noticeably straight trunk, tapering gently from the ground. The bark starts off a very smooth light grey with fairly obvious white horizontal lenticels, which are a little cherry-like, but the bark becomes much more rugged as the tree ages.

It is a little surprising to find the tree growing here in West Kent, as it is said to need an average of 730 mm rainfall, and therefore to do much better on the western side of the UK where the rainfall can be well above this.

The wood is said to be quite a warm reddish-brown, fine-grained and ornamental enough for furniture, etc. Potentially an interesting forestry tree, in the west and also in Scotland providing seed lots with a sufficiently hardy provenance are used.   

The Common Horse Chestnut, Aesculus hippocastanum

https://theecologist.org/2017/may/12/demise-horse-chestnut-tree-armageddon-or-scaremongering

https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/1365-2745.13116

The Japanese Horse Chestnut, Aesculus turbinata, not.

Owen Johnston has identified a tree in the Horse Chestnut collection at the front of Hadlow College as the Japanese Horse Chestnut, Aesculus turbinata.. I think that this species is really very similar to the common Horse Chestnut, A.esculus hippocastanum., but the tree identified is fairly certainly the Yellow Buckeye, Aesculus flava.

The pear-shaped(really?) fruit of pure Aesculus turbinata is without spines, although they may be warty and the hilum on the conker is much bigger than Aesculus hippocastanum, making the species easy to distinguish when with ripe fruit. There is a full range of hybrid types!

The leaves of Aesculus turbinata are a little more regularly toothed, narrower, to 40 cm, not so obviously wide above the center, and tapering more gradually to the tip. The leaflets do not have a petiolule, unlike the Indian Horse Chestnut, Aesculus indica. The underside of the leaves should be glaucous green, with hairy veins, rather than yellow green. The autumn colour is supposed to be "distinctive" - reddish perhaps. The buds are viscid. 

The flower spikes or panicles are bigger (25 cm x 6 cm) but supposedly rather less obvious, a little more "hidden" than those of Aesculus hippocastanum, and produced about 2 - 3 weeks later. The petals should be cream with red spots. 

There are two trees there at the moment, one bigger, one smaller, both in early stages of leafing out - could one of them, or both, be it?? Neither tree matches Aesculus turbinata. My best guess is that the biggest tree is Aesculus flava, and the smaller is Aesculus indica.

The interesting tree at Luddenham Court, visited in the second half of April, next to the Red Horse Chestnut, Aesculus x carnea, is quite a possibility. The leaves are already quite tattered, a bit like a Common Horse Chestnut, which I haven't seen in the on-line pictures much. however the key feature I was looking at was the absence of spines on the first stage of the developing fruit, so I was pretty sure it was not Aesculus hippocastanum.  The leaflets did not have red petiolules, so it could not be Aesculus indica, the other major possibility.  However the flowers were very white rather then cream, rather pretty and bright, and the leaves did not look that much different from Aesculus hippocastanum. Also, I didn't note any obvious glaucous underside of the leaves.