A Visit to a Real Beer Analysis Lab

Yesterday  I got the opportunity to visit an actual beer analysis lab where the lab analyzes a commercially made beer's bacteria count, pH, SRM (color), conductivity, IBU's, density and alcohol %. It was an amazing opportunity where I got to be hands on and my previous biology and chemistry class lab experience helped me not feel like I was walking into a nuclear reactor and asked to fix it.

The reason for these beer analysis labs is for smaller breweries who do not have internal labs to quantify their beer's quality control to compare to sensory analysis and also confirm that the labels they are sticking on their bottles for retails sales also match up with whats ACTUALLY inside the bottle.

Bacteria Counting
Because of the yeast plating I have been doing, I figured to count bacteria we would be using more petri dishes. WRONG! 3M makes disposable gridded pre-made media that look like sticky notes, except the media on the sheets can be tailored to whatever it is you are trying to grow or identify.

I first unboxed commercial samples of beers that came in from all over the country. Then I went through and labeled a corresponding 3M sheet to the sample given. I then micro pipetted out a small volume on to the 3M sheet, put the cover down and let it bleed out. These sheets will then incubate for 2-5 days to allow bacteria to grow. If it does grow, it will show up as little red dots. The sheets that have red dots will be cell counted to get an approximate bacteria count for the batch.

For all subsequent tests, all beer samples were portioned out into new beakers and labeled again. Small samples were taken.

I also got to use this handy dandy tool which makes pipetting SO easy! It uses vaccum pressure to pull up an amount and push out an amount. Where has this been my entire life?!?

IBU's
To measure the IBU's (International Bitterness Units) in beer, you use a liquid extraction technique described HERE in great detail, but to sum it up, alpha acid (humulone) is insoluble in water. That is why you must boil your bittering hops for at least 60 mins to get any bitterness utilization and isomerize the humulone. But to measure IBU's after the beer has been packaged you must extract those alpha acids OUT of the beer. You take HCL to break acid out, then use 2,2,4-trimethylpentane (isooctane) to snatch the broken out alpha acids to then test.

trimethylpentane on top, beer on the bottom prior to shaking

This sample with the HCl and trimethylpentane are then shaken for 15 minutes to emulsify, then centrifuged, then placed into a spectrophotometer where are wavelength of light is used to measure its absorbance. A final calculation is made with this value, and voila, IBU's!

This is the spectrophotometer

SRM (color)
Just like the IBU's, the spectrophotometer was used, however no preparation was needed of the beer sample, other than off gassing the beer and filtering it if it was cloudy, such as an wheat based beer.

Density & Alcohol %
This is where the expensive equipment was used. Density and alcohol % of an alcoholic beverage are linked, as long as you have pure ethanol in pure water. It is a simple D = M/V equation. However with beer, you have other solutes in the solution. So you need to separate the ethanol from the beer through distillation (which is time consuming) OR! you can use a super expensive Anton Paar Alcolyzer and Density meter that will report all sorts of fun facts such as the alcohol %, but also calories, extract, degree of fermentation, and others!

This machine setup and automatically churn through up to 20+ samples in one go

Conductivity & pH
Finally the last tests run were conductivity and pH. pH is an obvious choice to measure, since pH really influences how a beer is fermented and of course its flavor (think sour beers and its lower pH). However! have you ever considered testing the beer's conductivity? The reason to do so is to check the salts present in the beer and compare that to your water chemistry used in the brew. Do they compare? Are you getting any off flavors that you think are due to too low or high salts? This test might help with that.

Overall Takeaway
Considering I am very early on in my school career for fermentation sciences, this has A) given me a heads up of what to expect. B) A possible career field within fermentation to pursue and C) more general knowledge of the field I am heading into.

Tree Bark Yeast, Part 2

Introduction
This is part 2 of a hopefully 3 part series of catching, isolating, and using a yeast caught from a tree's bark. In part 1, I attempted to catch a yeast from an oak, apple, and pine tree. The pine tree yeast never panned out, but I got possible catches on the apple and oak tree bark.

Objective
To determine what species of yeast I caught on the apple and oak tree bark.

Prediction
The yeast caught on the apple tree is a Brettanomyces strain. The yeast caught on the oak tree is a Saccharomyces strain, possibly S. Paradoxus.

Materials Needed
DME
Agar Agar
DI Water
Yeast Nutrient
Bromocresal Green
Cycloheximide
Pietri Dishes
Alcohol Burner
Loop
Santization Liquid

Procedure
Plating
I used agar DME pietri dish plates, which I found a good recipe HERE, that uses 35g DME + 2.5 agar per 400ml of DI. I added a pinch of yeast nutrient. I also added the Bromocrescol Green. Bromocresol Green is a bacteria inhibitor. This should allow only yeast strains to grow and eliminate the growth of bacteria.

Brett should be able to metabolize the bromocresol while the saccharomyces cannot. By metabolyzing, that means if a strain cannot metabolize bromocresol, the colony will remain green. If it can metabolize, the colony will grow white. Some strains of sacchromyces can metabolize bromocresol, so it is possible to get a false negative or a false positive. My research has shown that this metabolizing factor is not a very solid determination of a Brett or Sacch strain. This part of the experiment isn't that crucial if I don't get a mix of green and white colonies. Bonus points if it happens.

Once enough colonies had formed for both the apple and oak plates, I selected 3 good looking colonies from each and replated those colonies onto new Bromocresol Green plates. This allowed a backup of the colony, as well as allowing me to select even newer formed colonies to go onto the final plates to confirm or deny that these strains are Saccharomyces or Brettanyomyces.

These final plates contain cycloheximide. Cycloheximide inhibits saccharomyces growth, yet allows brett growth, thus this will be my final determination of Brett or Sacch strains.

Here is a flow chart of what I am attempting:

Analysis
Here are the final images of the above's worflow diagram, but in picture form of each plate. Some of the plates are a little old when I took the picture, but they worked flawlessly when they were at their peak.

Apple Tree	Oak Tree

Conclusion
My prediction was mostly correct! The apple tree did in fact have a brett strain in it, as well as two sacch strains. The oak tree was 100% sacch.

What kind of brett and sacch strains did I catch? There is where I start to tread into unknown territory. Unless I sent out a sample to be DNA fingerprinted, the only other option is to do my own PCR and electrophoresis (which I have experience doing on human hair) to fingerprint its ribosomal structure, but I don't exactly have easy access to that kind of equipment nor primers. Future goal?

As for the bromocresol indicator to determine brett vs sacch? Not a worthy tool, BUT! totally worth using to eliminate bacteria. It feels really good to know that there is no bacteria growing on your plates and you only have yeast growing.

Plating "Control"
Because this is my first time doing all of the above, I also did 2 "control" plates. One was all filled with the regular agar, the other with the bromocresol agar. I then looped a commercially bought pure strain of each of sacch, brett, and lacto into each third on each type of plate. This was to see if the bromocresol does in fact inhibit the bacterial growth (it does) and to see what brett and sacch should look like.

Continuing onwards....
For now, its time to move back towards the sensory analysis side of the experiment. I have created 6 75 ml starters, one for each 1/3 of the plate for the apple and oak isolation plates. Now that I have pure strains, hopefully I can get some better ferments than the initial ferments in part 1 of this experiment, where there is no bacteria competing with my live strains to form and off aromas and pellicles.

Tree Bark Yeast, Part 1

Introduction
Wild yeasts are all around us. To think that waaaaaay back when, sugary liquids just "magically" fermented on their own, people had no idea that these tiny single celled microorganisms called yeast were actually doing all of the work.

Where do these wild yeasts exist? They are floating around the air, living on all sorts of things, and even inside things, like wasps. After random internet searching I stumbled upon an article stating that a cousin of Sacchromyces Cerevisiae, S. Paradoxus actually lives on the bark of Oak Trees. Oregon happens to have its own species of Oak, Quercus Garryana. And one person somehow got a pitch of this yeast and made a beer out of it with some possibly promising results. So why not try to catch my own, and or something else along the way? Because finding an awesome usable wild yeast, is like instant bragging points in the homebrewing world right? Kind of like Beer Cred?

This post is Part 1 of hopefully a 3 Part series. Part 2 will take a hopefully usable strain, and isolate that strain, then grow it up into a sizable pitch for a 1 gallon beer ferment. Part 3 will be using that pitch to make the 1 gallon batch of beer.

Objective
To catch a usable strain of yeast off of a tree, and use sensory analysis to determine if the yeast is USABLE.

Prediction
I will catch a USABLE strain of yeast off of a tree that will eventually ferment a delicious beer.

Materials Needed
DME
Water
Yeast Nutrient
Scraping tool
Vials
Sanitizer solution
Heating pad

Procedure
Finding wild yeast
According to the article above, wild yeast like to live in the bark of trees or in the "extracant" of trees, aka, liquid matter that a tree will release, like pine tree sap. Here in Oregon we have primarily decidious, coniferous and fruit trees scattered all over the place. So finding a tree isn't hard, its understanding WHERE on the tree to look for a location.

I already decided to look at an oak tree, but I also wanted to try pine tree sap and one other tree type to get 3 totally different samples.

Capturing wild yeast
For the oak tree,  and for everyone I examined, the bark was pretty uniform, and even if there was a rough spot, I didn't notice any extracant. So I took a knife and carved a hole in it and grabbed small samples of every layer of bark until I hit a nice depth.

For the pine tree, I found one that had emitted pine sap pretty recently, and also awhile ago. I grabbed small samples of old and new.

For the 3rd, this was a total awesome surprise: I have a 20-30 year old (estimated) apple tree in our back yard. A large branch was cut off some years ago and the cut location scabbed over. It had just been raining (welcome to Oregon!) and I noticed yellow droplets on this scab. I also noticed a number of fruit flies buzzing around this very location. Fruit flies = sugar! Sugar = yeast living near! So I scraped up this yellow droplet material and some "stuff" growing on the scab as my sample. I have high hopes for this one.

Fermenting with wild yeast
To get my samples to actually ferment something, I made up a 1.040 sample of wort using water, DME, and a pinch of yeast nutrient. I boiled the solution, chilled it and filled up used White Labs vials that had been cleaned and sanitized. Each sample was added to its own vial of wort. All three samples will sit in a glass to remain upright and this all will be placed on a heating pad and covered to maintain a constant 72* temperature to allow favorable fermentation temperatures.

I am going to give each sample 10-14 days to do their thing and grow a large enough sample that I can throw it onto a stir plate if the sample it fermented smells good or it will get tossed if it ferments something that smells like either poo, puke, or garbage.

Sensory analysis will be done in the morning and evening to record the appearance and aroma of each sample. For appearance, bubbles forming, cloudiness, and a hissing of air when opening a vial will determine how the ferment is going. For aroma, looking for a fruity, bubblegum, and other favorable smells will be good, whereas poo, puke, and garbage smells will tell if the ferment is going badly.

Observations

			Appearance			Aroma
Date	Time	Apple Tree Scraping	Oak Tree Bark	Pine Tree Sap	Apple Tree	Oak Tree	Pine Tree
11/28/16	PM	Clear wort, some specs of material, settled wort trub	Clear Wort, with some bubbles from agitation, settled wort trub, oak tree stuff	clear wort, settled wort trub, pine sap	Wort	Wort	Wort with pine from sap
11/29/16	AM	bubbles starting to form	same	same	doesn't smell like wort anymore, clearly something is fermenting	Still wort smelling	still wort smelling with the pine sap
11/29/16	PM	more bubbles, slight cloud	same	same	smells like Brett	same	same
11/30/16	AM	very cloudy	First Signs of Fermentation, some bubbles	First Signs of Fermentation, some bubbles	stronger brett funk, some cheese, corn	neutral smell	wort smell is gone, neutral smell with pine
11/30/16	PM	still very cloudy	Getting cloudy, but not as cloudy as pine	Getting cloudy	super strong brett	some brett smell forming	some brett smell forming, with pine from sap
12/1/16	Midday	no change	cloudier than pine, lots of bubbling activity. Large hiss from opening cap	no change, doesn't look like much is happening	still super cheesy and bretty	gaining more cheese and bretty, getting similar to apple tree	no change
12/2/16	AM	More bubbles, still as cloudy, pellicle has started to form	LOTS of activity. Large hiss when opening, causing debris @ bottom to shoot to the top, very cloudy	maybe slightly cloudy? no signs of bubbles	not as cheesy, but still has brett smell. Calmed way down in terms of smell intensity	smooth out a bit in terms of smell intensity	no change
12/3/16	AM	hiss when opening; pellicle is getting thicker; still as cloudy	TONS of activity! opened up to HUGE hiss and then a layer of foam wanted to shoot out of the vial! stirred everything up pretty violently. What a krausen!	no change	slight hint of spice; but other wise still the same level of cheese and brett	slight spice, very very slight fruit, cheesiness has disippated some; brettyness has balanced	no change
12/4/16	PM	half the pellicle disappeared; still as cloudy	still a HUGE hiss and krausen and violent mixing upon opening vial, but not as huge as 12/3/16	no change	cheese has disappated; mostly bretty smell	clove and cinnamon are coming out the strongest with a slight fruit; still some bretty character	no change
12/5/16	Mid Day	pellicle is still there, but co2 bubbles have become trapped in the pellicle; still sight hiss when opening	eruption intensity has dropped by 1/2	mold is starting to grow; still no CO2 activity	same	same	no change
12/6/16	AM	same	eruption is a tiny fraction of the max a few days ago. Still some outgassing; things are calming down for sure; peak was reached on 12/3/16	no change	same	more bubblegum is emerging	same
12/7/16	Mid Day	pellice dropped	no activity of bubbles when opening; barely a hiss when opening	more mold	same; still brett characteristic	still bubblegum	same
12/8/16	Mid Day	no change	same	same	same	same	same
12/9/16	AM	same	pellicle is starting to form	same	same	same	same

Analysis
Of all three experiments, only the pine tree sap was a total bust. Nothing really happened. Is there some form of antimicrobial property that lives on pine tree saps that inhibits yeast from growing on it?

For the other two, apple tree and oak tree, both did something, and enough to pursue further.

For the apple tree, it took off the quickest, but didn't really reach an explosive level of activity. Its smell was more on the brett side of the spectrum. On the other hand, the oak tree took a little bit to get going, but once it got going, IT GOT GOING! In terms of activity, on 12/3/16 when I opened to inspect I had to quickly close it back up or else lose half the volume to a foam-over. Thats a good sign of fermentation! The oak tree's smell started off bretty and cheesy much like the apple tree, but eventually the brett and cheese vanished, only to be replaced by spice and fruit and bubblegum, a possible sign of a Saccharomyces strain.

Conclusion
Based upon sensory analysis, my conclusions are going to be more like predictions for part 2 of this experiment. I have no definitive way (at the moment of writing this) that the apple tree strain is a Brettanyomyces strain, and the oak tree sample is a Saccharomyces strain.

But because Brett tends to be more cheesy, barn-yardy, horse-blankety, and the apple tree smells like all of those things, that is why I am "think" it is a Brett strain. And because Saccharomyces strains can be spicy, fruity, and bubblegummy, I "think" the oak tree strain is a Sacch strain. Also because of my research that S. Paradoxus grows on Oak trees, perhaps I did catch S. Paradoxus on my first attempt?

Future experiments will tell...

Encapsulated Yeast Beads, v1.0

Introduction:
This was my first attempt at making encapsulated yeast and fermenting wort with them.

Objective:
To make encapsulated yeast and ferment two 3L batches of a beer. One is fermented with a traditional yeast slurry (the "control"), the other is fermented with the yeast beads. Do sensory analysis to see how much different each batch tastes.

Prediction:
Encapsulating yeast in alginate beads will ferment out a wort solution faster than a yeast slurry, and will allow for easy separation of the yeast beads from the finished fermented wort. No prediction on how they will taste (in terms of sensory terms), other than different.

Materials needed:
Tap Water
Distilled Water
DME
hops
sodium alginate
yeast (Safale-US05)
yeast nutrient
calcium chloride

Process:
Yeast starter
I use a 100g DME/1000ml tap water ratio. I created a 2L starter (200g DME/2L tap water) with a pinch of yeast nutrient. I pitched the dry Safale-US05 in the starter on the stir plate to get it going. I know you aren't supposed to do yeast starters with dry yeast, but I do anyways because it at least gets the yeast really primed and ready to ferment with a slower lag phase. I have no evidence yet if what I am doing actually helps. I guess I just want a larger yeast count? Anyways, this is what I did.

Wort creation
To make 6L (or about 1.5 gallons of beer) to be split into two 3L batches, I used 1.5 lbs of DME, and 1.5 gallons of water....in a 2 gallon pot. I had to split the wort up and put 2L into a smaller pot to prevent a boil over. I need to get a larger pot...

Once the larger pot was boiling, I added 6g of old Nugget whole leaf hops to give some bitterness. With one minute remaining in the boil, I added 6g of old Saaz whole leaf hops for some aroma.

I cooled the wort as rapidly as possible in an ice water bath in the kitchen sink.

Beads creation
(Backing up to the previous Thursday before explaining this experiment's bead creation...)
The previous Thursday I made up a 4% Sodium Alginate Solution (4g alginate/100ml distilled water). The solution was so incredibly thick I could hardly stir it into a consistent solution nor could my immersion blender even sink its teeth into mixing it evenly. So I added 100% more water to make a 2% solution. I was worried this was too thin. Before adding any yeast, I tested out the creation of beads in the 3% calcium chloride solution.  The 2% alginate created REALLY tough beads. My gauge of "toughness" was how much effort it took to crush a bead between my fingers. I figured the tougher the bead, the thicker the membrane and the slower the diffusion of the beads. So I made 1% and 0.5% solutions and tested each. 0.5% was too thin and would crush too easily, while the 1% felt like it had a sweet spot. It would still crush, but didn't feel fragile. My 2% "stock" solution had lots of bubbles in it at this point, so I let it sit because I was leaving for the weekend and had no time to conduct this experiment.

(And now for this experiment's bead creation...)
By now the 2% "stock" sodium alginate solution had no more bubbles in it. I had cleaned the yeast from my starter had about 200ml of pure yeast to use for this experiment. Because each batch was 3L of volume, I had no idea how much volume of beads to create. Because I needed some of the yeast to pitch as a direct slurry and the rest to make beads, I figured a 10% volume of beads would be a great starting point. This would be 300ml of beads. This would mean to create a 1% alginate bead, I would need to cut the 2% stock in half with yeast. So that means take 150ml of the yeast slurry and mix with 150ml of the 2% alginate to make 300ml of beads. The remaining 50ml of yeast would act as the slurry in the "control" batch.

With 300 ml of alginate/yeast mixture, it was time to make the beads. I used a brine injector without its needle to make the alginate/yeast droplets to drop into the calcium chloride solution. Because this was a 1% SODIUM alginate/yeast mixture, and because the reaction to create the beads is a replacement reaction where the sodium ions of the sodium alginate swap with the chloride ions of the calcium chloride solution, the sodium ion is going to be the limiting reactant, so as long as the calcium chloride solution is greater in volume and greater in chloride % than the alginate/yeast mixture, then the calcium chloride bath won't "run out" of chloride ions.

So now the process of creating the beads was mechanical at this point. Use the brine injector to suck up around 30+ml of alginate/yeast mixture, then slowly drop droplets into the calcium chloride bath. This took awhile to create 300+ml of beads. I also noticed that the closer to the calcium chloride bath the beads were dropped, the more tear dropped they were shaped. Whereas the higher they were dropped, the more spherical they were shaped. Spherical is better = more surface area.

I actually wound up with a little more than 300 ml, but it was ok. I figure the more the better.

Fermenting:
After filling each 1 gallon jug with the split wort, and after pitching the control with 50ml of yeast slurry and the other jug with the beads, I placed them on a plant heating mat with a temperature controller set to 72*F. A placed boxes from 3 gallon carboys over each jug to keep the wort from sunlight and to create even heating and preserve heating.

Data:
I collected data every 12 or so hours. I used a brix refractomer and relied on a correction calculator to correct fermented brix values as well as convert to gravity. The data is below.

	Brix				Adjusted Gravity
Date	Slurry	Beads	Beads Comments	Date	Slurry	Beads
10/31/2016 17:00:00	10	10		10/31/2016 17:00:00	1.04	1.04
11/1/2016 7:00:00	8.5	8	Maybe?	11/1/2016 7:00:00	1.029	1.026
11/1/2016 17:00:00	6	9.25	huh?	11/1/2016 17:00:00	1.014	1.034
11/2/16 7:00 AM	4.75	9.5	nothing is happening	11/2/16 7:00 AM	1.006	1.035
11/2/16 7:00 PM	4.75	8.5	something is happening, its making lots of micro bubbling noises	11/2/16 7:00 PM	1.006	1.029
11/3/2016 7:00:00	4.75	9	Lots of little bubbles and more popping bubble sounds. Not as clear	11/3/2016 7:00:00	1.006	1.032
11/3/16 5:30 PM	4.75	8	Slightly cloudy, very active	11/3/16 5:30 PM	1.006	1.026
11/4/16 9:00 AM	4.75	6	kinda of clearer? Lots of little bubbles. Humming away	11/4/16 9:00 AM	1.006	1.014
11/4/16 10:00 PM	4.75	5.1	Still lots of bubbling noises	11/4/16 10:00 PM	1.006	1.008
11/5/16 9:00 AM	4.75	4.95	Still lots of bubbling noises	11/5/16 9:00 AM	1.006	1.007
11/6/16 8:00 AM	4.75	4.5	A lot quiter. Very Clear. Done?	11/6/16 8:00 AM	1.006	1.004

This was taken about 12 -24 hours in.

Data comments:
Both batches' airlocks were bubbling within 5 hours. My first reading 12 hours in had me really excited as the beads had dropped further than the yeast slurry. However I think that was a false reading. The following 4-5 readings really gave no direction as to what the beads were doing. It wasn't until the evening of 11/3/16 that it was clear the beads were actually doing something.

Results:

The yeast beads worked. They didn't work as fast as the yeast slurry, but they worked. Despite working slower, they did attenuate further than the yeast slurry and kept a very clean ferment, as in no haziness in the ferment. The "control" batch had a classic heavy swirling cloudiness during ferment and is slowly clearing as the yeast flocculates. There was only a day or two of cloudiness in the beads batch, and this paled in comparison to the yeast slurry batch.

Finished Ferment. The clarity is amazing in the beads

Analysis:

I think I way over pitched the yeast slurry. Never seen a batch ferment out in 1.5 days. There was quite a lag with the beads, but once they got going, it actually fermented out the batch almost as quickly compared to the slurry, though not quite as fast. But what is the equal amount of yeast to be used in a comparison ferment like this? 50ml of 3000ml is 1.67% of volume whereas I used 5% of volume with the initial 150ml of yeast slurry for the beads. I am going to leave this as a question to be answered later.

It is kind of hard to tell, but I am not sure if 100% of the beads were fully submerged into the wort. This might mean that only the beads on the bottom (weighted down by beads above) were doing all of the work? What would happen if 100% of the beads were submerged in the wort?

Sensory Analysis:

Sensory Analysis	Slurry	Beads
Appearance	Super Hazy, almost like a Wit	Crystal Clear, Like a lager
Aroma	Very Floral, Almost Saison Like	A layer of floral has been stripped off. Almost chemical
Flavor	Ok	Not great
	Bitterness is softened	bitterness is HARSH
	Some breadiness from the yeast	Chemical smell comes through as chemical flavor
Mouthfeel	Soft, chewy	Still soft, slightly thinner, but not much
Overall	Terrible recipe to begin with, but better than the beads	Terrible recipe to begin with, and a terrible beer. NOT good at all!

Yes, the recipe I used for this beer was terrible. I guess I added hops to make sure other things didn't start growing in there? I guess (and this was a happy accident) I never thought much about how the bitterness and aroma characters would carry through. When you only ferment with a yeast slurry, that is all you know in terms of how a beer will ferment. It is not until you have a side by side like this when you realize just HOW bitterness and aroma can be affected simply by the ferment.

I also took for granted just how much flavor yeast will add to your beer. The yeast beads simply had ZERO yeast character (which was kind of the point...) but in comparison it gives you an awesome perspective on HOW yeast adds flavor to a beer.

Final thoughts:

Going forward I should have harvested the beads as soon as fermentation was done and immediately added them to the next new batch to see if they are "primed" and ready to immediately start fermenting. I should do this side by side with some brand new beads. And I should do two versions of the beads (so 3 one gallon batches side by side), one with the same % of alginate, and one with 1/2 the % of alginate to see if a thinner membrane of the beads will allow quick diffusion of sugar to alcohol.

As for actually using this method for making a good beer? The sensory analysis shows A) don't make a shitty beer to begin with. B) Don't even think about making a beer where you want the yeast to add any character to the flavor: Hefeweizen, Saison, pretty much anything Belgian, the list goes on.... C) I guess the only thing "good" about the beads beer was that the aroma came through and was terrible, but not as terrible as the flavor? I am stretching here. Maybe a NE IPA could be a candidate? I am cringing...