Microbial Colonies of Pu’er Tea – new research

I’ve been talking / teaching / writing about the importance of the origin and cultivation of the microbial colonies within Pu’er tea for years,as they’re responsible for everything from “aging” rate to flavor profiles. (Reminder: all tea ages at the same rate of 1 year per year)

Yet, outside of the Institute, I’ve gotten some pushback;
including some crazy counter arguments like  pu’er is never actually fermented, or that there is no effect on colony strains from the factory / plantation / forest, or that the steaming process kills of the microbial colonies during compression (it doesn’t).


This new paper (June 2016) does a good job of showing just how important the microbiological colonies and inoculations are in determining arability, flavor profile, and quality of Shou Pu’er.

The Microbiome and Metabolites in Fermented Pu-erh Tea as Revealed by High-Throughput Sequencing and Quantitative Multiplex Metabolite Analysis



Pu-erh is a tea produced in Yunnan, China by microbial fermentation of fresh Camellia sinensis leaves by two processes, the traditional raw fermentation and the faster, ripened fermentation. We characterized fungal and bacterial communities in leaves and both Pu-erhs by high-throughput, rDNA-amplicon sequencing and we characterized the profile of bioactive extrolite mycotoxins in Pu-erh teas by quantitative liquid chromatography-tandem mass spectrometry. We identified 390 fungal and 629 bacterial OTUs from leaves and both Pu-erhs.
Major findings are: 1) fungal diversity drops and bacterial diversity rises due to raw or ripened fermentation, 2) fungal and bacterial community composition changes significantly between fresh leaves and both raw and ripened Pu-erh, 3) aging causes significant changes in the microbial community of raw, but not ripened, Pu-erh, and, 4) ripened and well-aged raw Pu-erh have similar microbial communities that are distinct from those of young, raw Ph-erh tea.

Twenty-five toxic metabolites, mainly of fungal origin, were detected, with patulin and asperglaucide dominating and at levels supporting the Chinese custom of discarding the first preparation of Pu-erh and using the wet tea to then brew a pot for consumption.

Kilns for Firing a Yixing Teapot

NOTE: this post is out of date. Those Interested in Yixing Teapots are suggested to read An Introduction to the Art & Science of Chinese Tea Ceremony: Yixing Teapots – Knowledge, Connoisseurship, and Technique

There were two types of kilns historically used in Yixing: Snake Kilns and “Modern Tunnel” Kilns. Later, gas and electric kilns came to be used instead.

Snake kilns were used throughout China for thousands of years, and the only type of kiln in use up to 1957. Snake kilns are simple, single-column climbing kilns made of clay and brick with a frontal firing chamber, air and fuel ports along the side, and a tall chimney at the far end. The kiln is set on a slope with the chimney at the highest point, which causes a consistent oxygen flow due to the heat rising up the slope and out of the chimney. These kilns could be up to 60 meters long and burn for over a week (though they would have been shorter and faster for yixing firings).

burn baby burn!

Continue reading

Basic Verification of Yixing Teapots

NOTE: this post is out of date. Those Interested in Yixing Teapots are suggested to read An Introduction to the Art & Science of Chinese Tea Ceremony: Yixing Teapots – Knowledge, Connoisseurship, and Technique


calligraphy on a teapot; flowing liquid poetry; I want more tea to speak to me

Yixing teapots have been in nearly continuous production (wars being the primary cause of discontinuity) since the late-Ming or Early Qing dynasty.

The increase in price and rarity of real yixings of known provenance has caused a ripe market for fakes – often an expensive mistake for new and experienced practitioners alike. Simple fakes are often made from clay other than Yixing, and are not particularly hard to identify.

The better “fakes” can be real Yixing – but from the wrong year or maker. Most yixing teapots were made by students and apprentices of masters, who learned to make teapots by copying the designs of the masters before them. Masters from previous dynasties would have both later masters and their students copying their designs – “fakes” existed as early as the Qing Dynasty, with potters copying master made pots from the Ming Dynasty. Students and apprentices copied their masters’ designs using the same tools, clay, and kiln. Thus, most master made pots would have been copied at least 10 times by students in training, using identical clay and equipment.

Are all of these teapots fakes? No – that would be a poor definition of “fake”…. Do yixings need to be master made to be real? No – they simply need to improve the tea you decide to pair with it to be used within GongFu.

Thus, even if one knows how to determine the composition of the clay (yixing vs. not yixing) by hand or through an analytical test, determining the age and provenance is still a challenge (for example, you can’t use radio-carbon dating on clay….).

A practitioner of tea who wishes to purchase Yixing for GongFu must learn a set of methods to separate real from fake. This first post will explain.

Continue reading

Processing of Yixing Clay

NOTE: this post is out of date. Those Interested in Yixing Teapots are suggested to read An Introduction to the Art & Science of Chinese Tea Ceremony: Yixing Teapots – Knowledge, Connoisseurship, and Technique

The raw yixing dirt needs to be processed and refined before it can be fired into a beautiful, usable, teapot. The clay starts as a dusty, soft, colorful dirt and goes through 8 processing steps that can take up to 50 years to complete – the longer the clay is aged and rested, the higher the quality (and the less likely it is to break in the kiln from uneven shrinkage caused by impurities).

This process list only applies to higher quality yixing clay. Obviously, lower quality yixing clay (and fake or fraudulent clay) cuts corners and the aging and resting process.

The Process:

  • The clay-dirt is left out in the elements (for a long time) to dry. The particle size of the uncompressed clay is quite small, and the clay should be spread out over a wide area.
  • The drying process (and mallets) breaks down the clay into even smaller particles. This makes it easier to sift and remove impurities.
  • The Yixing continues to rest and soften for a minimum of 2+ years (usually about ~3 years).
  • The clay is dry milled into sand. (this is where the term purple sand for Zisha Yixing comes from)
  • The sand is sifted with water down to 30 microns (the laborers where respirator masks). This continues to remove impurities and pieces of yixing that haven’t broken down.
  • The artist or yixing master blends refined sands from multiple areas. The mixture is usually secret, and only passed to apprentices in their final level of training (at least in the past… the nationalization of the yixing industry changed this, and now… some yixing artists have returned to using secret blends, but how secret they are is questionable). In the past, many yixings we’re made from clay from single mine – blending was rare. Now nearly all yixing teapots are made with blended clay.
  • The clay is mixed and churned to the right consistency. Water is consistently skimmed off the top to continue removing floating impurities.
  • The clay is kneaded, hammered, and thrown into a block. It is now ready to be used for ceramics.

Remember: the clay is the single most important attribute to consider when purchasing a yixing teapot. You should be purchasing the yixing teapot for its positive affect on the flavor profile of a single type of tea – and the clay is responsible for 99% of that effect. The shape and size matter much less than the purity, quality, and firing of the clay.

Projects in Machine Learning and Artificial Intelligence on Tea

What could a machine possibly learn about tea? And why would any AI want to learn about tea? We can barely find any humans to study tea!

I taste therefore I am.

My company uses machine learning and narrow-band artificial intelligence to understand what people taste in complex food and beverage products – the AI we’re building probably won’t take over the word any time soon, but if it does, it’ll just make everything taste better.

So why AI?

AI is a bad term for a wide range of research and capabilities; our narrow-band AI can be better thought of as automated machine intelligence to identify and control for subjectivity in human sensory data. Our goal is to strip away the influences of age, sex, race, socio-economic status, past tasting experience, first language, and smoking habits from what people claim to taste in a product, and arrive at the underlying chemical composition of flavor active compounds – without any lab equipment.

Once we do that, the real fun and intrigue begins. For example, we use that data not only to build real-time quality control monitoring for flaws, taints, and contaminations in beer – but also to determine what percentage of the population will be able to taste any quality control problem and the overlap with the producer’s target customer demographics. In bourbon, we’ve developed flavor-profile optimization strategies in the production process, and have helped producers predict the optimal barrel and bottle aging, removing a lot of the guess work from the process.

What projects do I have in mind for tea?

Continue reading