Posted by: danlwarren | August 27, 2010

Microscope setup for field sperm motility study

One of the things Teresa and I are working on in Curacao is a study of sperm motility in bluehead wrasses (Thalassoma bifasciatum).  I’ll talk about why that’s interesting eventually, once we get some of the results out.  One of the hard bits of doing this study is that blueheads don’t manufacture sperm in captivity – we figured out very early on that they just shut down production after a couple of days, and fish from the pet trade therefore have no motile sperm to study.

Obviously, we need to be able to measure sperm motility in the field.  There are commercial setups out there that you can buy for sperm motility, but they have several disadvantages: they’re not very portable, they’re expensive as all-get-out, and most importantly they’re (last I checked) not very flexible in the types of video you can analyze or the timing of the analysis windows.  It’s been several years since I played with a Hamilton-Thorne system, but at that time you could only specify an analysis window by manually hitting “start” and “stop”.  That’s worthless for my study because of the short motility period of bluehead wrasses – their sperm are motile for a very short period (~15 seconds!), and those little suckers are FAST. I need to be able to analyze very precise time slices post-activation, and the HT system just couldn’t do that last time I checked.

So, in the great scientific tradition, I bodged something together.  This setup is the fourth such that I’ve put together, and is by far the nicest.  Earlier versions involved CCDs and video capture direct to the laptop, but the data produced by those setups was nowhere near as nice as what I’m getting with this arrangement.  So, without further ado, here we are:

The first thing we need is a microscope.  I chose a fairly cheap model from AmScope for a couple of reasons: theft is rampant in Curacao and I don’t feel like losing a really nice scope, for one.  More to the point, though, I’m setting this up out of my own very limited funds and AmScope makes astonishingly nice hardware for the money.  Certainly sufficient for my needs.  This particular scope is their T400A-30W-DK model.  They have another model that looks better on paper, as it has an adjustable photo port.  However, the adjustable port is only adjustable within a range of distances that vary from “useless” to “uselesser”: the focal distance is only suitable for cameras where the CCD is actually down inside the photo port, and I didn’t own a single CCD that worked.  In fact, I don’t know of a good video camera that does have a CCD that will fit in there.  The 400A can also fit into this kickass aluminum case:

Which the other model can’t.  Just look at that case.  I want to handcuff it to my wrist and walk around with sunglasses on.  It also does a great job of keeping the scope in working order despite being kicked around aiports, fondled by security guards, and bashed into all sorts of things (including my face on the return trip).

Okay, we’ve got a scope and a case, we can look at some sperm.  However, what we really need is some really good video.  As mentioned above, I was initially getting video to my laptop using a variety of CCD cameras and a USB video capture device.  This works, but there are several issues to deal with.  For one thing, the resolution and framerate of most of the affordable CCDs is awful for sperm motility.  Sure, you can buy an HD CCD if you want to shell out the dough, but then you have to buy an HD capture solution for your laptop and those are (a) hard to find and (b) damned expensive.  The low-res solutions are also usually putting out interlaced video, which is problematic for sperm motility analysis – it adds uncertainty to the motility parameters if you use the video raw, and all of the deinterlacing programs I tried led to positional artifacts that were worse than the interlacing itself (e.g., I’d get a nice deinterlaced picture, but the sperm was jumping back and forth between frames).

On this trip we decided to leap off of the purpose-built CCD train and try something off the shelf.  We’re taking advantage of the fact that Canon’s new T2i DSLR camera allows users to capture hi-def video.  The T2i is just an awesome camera all around, but what’s particularly nice for our purposes is that you can buy decent Canon EF -> microcope photo port adapters on Ebay for about $100.  By attaching the camera to the adapter and then to the scope, we can shoot hi-def video at a better framerate (720p, 59.94 fps) than the CCD solutions with less fuss and no f*&%ing video capture hardware.  Here ’tis:

But there’s one more issue to deal with.  This microscope head does not allow users to use the eyepieces and photo port simultaneously.  But hey, not a problem!  The T2i has an AV out, and we can use a really cute little off-the-shelf portable DVD player as a video monitor.  This is the Sony DVP-FX950.  It has a higher screen resolution than most portable DVD players, as well as an RCA video input jack.  Here it is, waiting for me to shoot some video:

We are getting really excellent results with this setup.  The one drawback compared to my old video capture systems is that I don’t have control over the filenames at the time of recording.  Because of this, I have to verbally announce the identifying information for each take so that the camera’s microphone will pick it up.  That means I’m going to have to go back and rename about 500 video files before I can start analyzing data, but it’s a small price to pay.  Speaking of small prices to pay, here’s the rundown for the whole rig:

Amscope microscope model 400A-30W-DK:  $419

Aluminum case: $90

Canon T2i camera: $900

Canon EF->photoport adapter: $95

Sony DVP-FX950 portable DVD player: $140

This comes to a grand total of $1644.  While that’s a significant dent in my pocketbook, it’s about 3% of the price of one of the Hamilton Thorne systems which, last time I checked, were going for something like $50,000.

Obviously there’s something missing here – the HT systems aren’t just video scopes, they’re analysis packages with software and everything!  Never fear, the amazing contributions of the open source scientific software community will come to our rescue there.  But that’s another post for another day.



  1. Are there any sample images available? I have a t2i and an amscope, and I am looking for a decent adapter for the camera.

    • I don’t have a picture with the T2i, but here’s a picture I took of some wrasse testis using my old XSi and the same adapter and scope:

      These pictures were just sorta snapped to make sure everything is working. Better focus and contrast are easily possible. If I get a chance this week I’ll try to snap a picture with the T2i.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s


%d bloggers like this: