China 2014

Pam travelled to Yunnan, China in the spring of 2014 to hunt for Primulas. Click image for more...

Primula Rediscovered

Primula bracteata and Primula bullata are found in their type locations after 125 years.

Near Lhasa, Tibet

How do you tell the difference between P. tibetica and P. fasciculata?

Primula ambita in the Wild

The first ever cultivated plant caused a stir at Chelsea earlier this year.

New Primula Book

The latest Primula book is a revision of the 106 species of Primula found in India.

A Primula from the Miocene - P. riosiae

The Miocene is a geological period extending from 23 million to 5.3 million years ago. It is during this time period that the apes arose and humans split to become their own lineage, open grasslands became more prominent and by the end of it almost 95% of modern seed plant families existed.
Evolutionary studies about our own primitive human ancestors makes the news with each exciting discovery, but behind the scenes, the evolution of the Primulaceae is also the focus of ongoing studies. Though genetic studies help us understand the relationships between species, other characteristics, such as seed morphology can be used to help differentiate plant species and to group related species. These days imaging seeds is easily done with clear results. See my Blog post Bringing Seeds into Focus.
Seeds are a goods means of exchanging plant material with like-minded enthusiasts worldwide. Many plant societies offer seed exchanges for their members including the American Primrose Society, Scottish Rock Garden Club, Alpine Garden Society and North American Rock Garden Society which all list Primula in their seed lists. But those are seeds of plants growing today. What would a Primula look like growing in the Miocene?
Primula riosiae from the original description
That is in part what Primula evolutionary studies hope to uncover but we can peek into the past and seed what a Primula seed looked like. The image above is the result of a Paleocarpological study done from samples collected in a lignite mine at Berzdorf, Germany which were dated to the lower and upper Miocene. Three seeds were found and they were then described by Alexander Czaja as a new species called Primula riosiae after Mrs. Anabel Rios in a paper titled "Paläokarpologische Untersuchungen von Taphozönosen des Unter- und Mittelmiozäns aus dem Braunkohlentagebau Berzdorf/Oberlausitz (Sachsen)" published in Palaeontographica Abteilung B Band 265 Lieferung 1-6 (2003), p. 49. The description is brief, but it is noted that the seed shows cells on the surface characteristic of many modern Primula seeds and that the seed bears a resemblance to Primula ruprechtii Kusn., which is considered a synonym of Primula elatior. Two recent papers which have used P. riosiae in their studies are : "Heterostyly accelerates diversification via reduced extinction in primrose" by de Vos et. al. and "Phylogeny and biogeography of Primula sect. Armerina: implications for plant evolution under climate change and the uplift of the Qinghai-Tibet Plateau" by Ren et. al.

Primula rebeccae A.J. Richards a synonym for Primula tenella King

Primula tenella was first discovered in the Chumbi valley, Tibet and was described by King ex Watt in Journal of the Linnean Society. Botany 20:13. 1882  (with a drawing) and also Hooker from Flora of British India 3:492.

Chumbi Valley - Reproduced by permission of Durham University Library and the Bentley Beetham Trust
W.W. Smith in The Genus Primula: Section Minutissimae. Trans. Bot. Soc. Edinb. 33:227-266. (1942),  says P. tenella was first collected in 1878 and again in 1879 in the Chumbi Valley, S. Tibet, by Dungboo, Sir George King’s collector. King recognized it as a new species and on the type (Dungaboo, anno 1878, in Herb. Calc.) has himself written the name giving the locality as Goop, 13 miles from Phari.” The holotype is presumably still at CAL, now Botanical Survey of India, Howrah, but there is no image of the type online.

At the Royal Botanic Garden Edinburgh is an isotype of a single plant with no location information or date, E00024523 and a second sheet collected by Dungboo in 1879 from Seain Chu(?), Chumbi, not available online. In Kew herbarium, there are two sheets online, K000639442, and K000639443. In Paris there is one sheet, P04907201The exact location of Goop is unknown, and access to the Chumbi Valley is restricted, so images from the approximate type location are not available.
Primula rebeccae from the original description
In 2000, Sabine and Georg Miehe found a Primula species growing near Jangothang, Bhutan (27°46'3.01"N 89°20'4.66"E) "at about 4000m and was confined to shady crevices amongst boulders in the vertically river-cut precipice between the flood-plain of the Pa Chu River and the adjoining river terrace. Here it grew in gregarious groups and clumps in a sandy-silt matrix on moss-dominated banks." The Miehe's also found similar plants higher up in the nearby Tso Phu Valley. This species was subsequently described by A.J. Richards as Primula rebeccae in Plantsman n.s., 3(1): 54. 2004. and the holotype is in Edinburgh, E00180782. The type location for P. rebeccae is about 36kms away from the type location for P. tenella.
Type location for Primula rebeccae

Subsequently, in 2008 and 2009, Margaret Thorne travelled to Jangothang and was able to photograph the population in detail showing this species to be highly variable. Notice the flower size, color, petal shape variation in the image above. Also notice how the leaves vary in shape with the exposure, back to front. See also the Species Gallery under P. tenella for more images.
Primula tenella by George Sherriff
In 1944, George Sherriff photographed Primula tenella (#16314) at the Lingshi Glacier, just 10kms away from the P. rebeccae type location. It is obviously the same species as that found by the Miehe's and even though the image is black and white, the pale eye of the flowers is apparent.

Primula caveana

In the original description of P. rebeccae, Richards compared the species to Primula caveana in Section Cordifoliae, a larger species which also grows in the same area. Differences given between the two species were the smaller calyx, filiform (thread like) pedicels, usually solitary flowers, emarginated petals and a pale eye and tube in P. rebeccae. However Richards did not compare the new species with P. tenella in Section Minutissimae, an obvious oversight. It is easy to see that P. rebeccae has the characteristic wedge shaped leaves shown in the original drawing with the original description of P. tenella and that it matches in everyway the (brief) original description of P. tenella.
Below is a map showing some historical collections and the location for P. rebeccae. 
Red- tenella, Blue - rebeccae, Yellow - caveana
(Yellow line is Tibet- Bhutan border)


Primula longipinnatifida

Primula blinii has numerous synonyms reflecting the variability of this species. One of the synonyms cited in the Flora of China is P. longipinnatifida F.H. Chen. This species was collected by T. T. Yü, NW of Wa-Erh-Dje, alt 2800m under woods in 1937. It was described in the article “An Enumeration of Primula Collected by Mr. T.T. Yü From Northwestern Yunnan” by Feng-Hwai Chen in Bull. Fan Mem. Inst. Biol., Bot. 1940. In the original description, Chen compared the specimen to Primula incisa Franchet which is a synonym of P. blinii. P. longipinnatifida was distinguished from P. incisa by having longer pinnatifid leaves, membranaceous in texture and by the smaller corolla tube. Chen cited Yü 6146 in the Herbarium, Yunnan Bot. Inst. (now KUN - Kunming Institute of Botany) but this is an error and the correct collection number is 6046.
Yü 6046 - Photo P. Eveleigh, courtesy Kunming Institute of Botany
Yü 6046 label - Photo P. Eveleigh, courtesy Kunming Institute of Botany

So why discuss this? In 2014 I visited the Kunming herbarium and looked at the holotype specimen of P. longipinnatifida. I was quite surprised that this species was so distinct, especially when compared to the type specimen of P. blinii. The leaves are so dissected they almost defy description. The texture of the leaves is like tissue paper and are not like those of P. blinii, rather they resemble P. runcinata which has leaves that aren’t quite so deeply pinnatifid and which has a different (racemose) flower arrangement.
Yü 6046 inflorescence detail - Photo P. Eveleigh, courtesy Kunming Institute of Botany

Yü 6046 leaf detail - Photo P. Eveleigh, courtesy Kunming Institute of Botany

Wa-Erh-Dje (now Waerzhai) monastery 28°31'53.51"N 100°55'9.06"E was one of three royal residences of the Muli King and was visited by Joseph Rock and illustrated in The National Geographic Magazine in July 1931. Rock described the location of the monastery as “3 days north of Muli”, Muli being the old site of Muli monastery not the present day town named Muli which is 50kms to the SE. Muli county is considered a Tibetan autonomous county within the Province of Sichuan and was an independent kingdom until 1950.  The present day county is split into three districts following the three most important monasteries: Muli, Kulu (Kangwu) and Wachin (Waerzhai). Travel in the northern part of Muli is restricted but foreigners are allowed access to the southern part though it is considered dangerous to travel there with its history of revolts and bandits. Some modern images exist of Waerzhai showing the old monastery ruins and the modern replacement. 

There is no indication by Yü how far to the NW of Wa-Erh-Dje the specimens were collected though the monastery is at 3300m and the collection altitude of 2800m is reached quickly in that direction as you descend to the river valley. Kunming herbarium has other specimens from Muli collected by Yü in the same year filed under P. blinii but unfortunately they are not imaged online so it is impossible to tell at present if they show plants closer to P. longipinnatifida or P. blinii (in the original sense).

This is one of those species that we need more information about and images from the wild showing population variation would go a long way to determining it's relationship with other species. We can only hope that someone will find this species again in the near future.