On four days last week, here is my day. I get to the lab a little after 8 AM. I unshroud the microscope, flick on the infrared light, and arrange bits and pieces on the lab bench, an operation that includes filling three small dishes with 3 mL of 1% congo red and making sure that the three special microscope-slide-chambers are ready. I fire up Larry Winship’s elderly Dell laptop, generously loaned to me for imaging, start Micromanager (the software that runs the imaging camera), open the relevant script, and organize the file storage directory for a new set of images. With that sorted, I go upstairs for a plate with seedlings. I choose the plate, place it a Styrofoam box between two hunks of aluminum that I store in the growth cabinet soaking up heat, and bring it all back downstairs to the microscope. I place the large square plate directly on the stage of the horizontal microscope and bring the first root into view. I align carefully so the root tip is right at one edge and rotate the camera if needed to ensure that the root is oriented horizontally in the image. I let the software rip. And by “rip” I mean the software tells the computer and camera to acquire an image every 10 seconds four times (i.e., four images), wait 30 seconds and repeat (Fig. 1). After the software finishes, I move the stage carefully to image the adjacent part of the root, because the entire growth zone is larger than the field of view. Taking care to set a new file name (“partB”), I run the software. I use a Sharpie pen to label the back of the plate with “1” and move the stage to bring a second root into view, repeat the image acquisition, write a “2” by that root, and finally do a third root. This completes task 1: live imaging. I will use these images later to measure the spatial profile of velocity within the root’s growth zone and from which I will get the elongation behavior. 

Next, grabbing the shoot with tweezers, I remove each seedling from the plate, carefully extracting the root out of the agar without breaking it. As the age of the roots gets older, day by day, removing them gets harder because the agar dries and lateral roots grow out and entangle. During this process, I keep the plate with the seedlings vertical to avoid gravitropic bending. With the root extracted from its agar home, I dangle its tip into a drop of water on a microscope slide chamber and lay out the rest of the root and shoot along the slide, often doubling back, because the root is longer than the slide. I make sure the entire root contacts water—otherwise the root will become welded to the glass. The chamber is formed from three layers of Scotch tape on each side of the slide, forming a channel for the root. Each chamber is numbered; root 1 goes in chamber 1, and so on. I add a coverslip and bring the three slides, along with my laptop, a USB adaptor, and an Allen wrench, around the corner to a conventional upright microscope. I attach the cable from the microscope’s camera to my laptop, open the software that drives the camera, and put slide 1 on the stage. I use brightfield imaging, this is the default method on a microscope, no special stuff required. I line up the root tip in the field of view, rotate the camera (that is where the Allen wrench comes in) so that the root image is horizontal, adjust the condenser iris to give a narrow depth of focus, adjust the focus to the top—epidermis—surface, snap an image (Fig. 2), move the slide to the adjacent position on the root, snap again, and repeat four or five times until I have moved into the position where root hairs have well and truly emerged. This completes task 2: imaging for cell file angle. I will use these images later to measure the pattern of twisting. 

I bring the slides back around the corner and place each seedling in a numbered dish containing 1% congo red (a solution that is more or less the color and density of blood) and place the dishes on a gentle shaker. Staining takes 30 min. I use this time to bring the plate with remaining seedlings and the aluminum blocks back up to the growth cabinet, tidy up the detritus from the first two tasks, copy the files from live imaging from Larry’s Dell to a USB drive, and catch my breath. After staining, I rinse the seedlings three times in water, five minutes each. With unbound dye rinsed off, I mount the roots for confocal observation. This involves again dangling the root tip into a drop of water on a (numbered) microscope slide chamber. Only in this case the slide chamber has just a single layer of tape. With the red-stained root tip lying on the slide in the water, I use a razor blade to cut off about a 1-cm-long tip segment and toss the rest of the root and shoot. I add a coverslip, gently, remove excess water with scraps of filter paper, and seal with nail polish (Maybelline clear). After 45 min for the polish to set, a period when I clean the dishes (used for staining and rinsing) and give the discarded shoots a decent burial, I take the slides up to the 8th floor confocal. This instrument—shrouded behind black drapes—is stuffed in an alcove at the end of someone’s lab bench. I turn on its three switches, start the argon laser and the instrument’s PC. I carefully organize the liquid crystal gear, which—between sessions— I have carefully stowed in a box out of the way. I plug in the  liquid-crystal controller to DC power and to a USB cable connecting the PC; and, I insert the slider into microscope, being careful to put it in upside down (or literally, right side up!). I start the confocal software and the liquid-crystal software, make sure they are coordinated, and put the slide with root number 1 on the stage. In transmitted light, I line up the quiescent center of the root (the origin for data from all three tasks) at the lefthand edge of the field; then, switching to fluorescence, I image adjacent sections of the root epidermis until reaching the region where root hairs have clearly emerged (Fig. 3). Because the various places along the root are stained to various extents, I have to fuss with the intensity of the laser and the gain on the photomultiplier. Because the root cells and files undulate, I have to fuss with the focus. To image the set of three roots, I need about 90 min. I copy the images onto a USB drive, turn off the laser and the confocal, and pack away the liquid-crystal gear. Finally, I convert the batch of confocal images to a form suitable for the polarization analysis. I could do this conversion step anytime but because it is super boring and takes about 30 min per set, I have built this step into the workflow—otherwise I will wind up with a baleful pile of non-converted images. Sometimes, this last step makes me want to scream but I have been doing it anyway. This completes task 3: imaging for cellulose orientation. I will use these images later to measure the orientation of cellulose and compare that to growth and twisting. 

Two features make this workflow intense. First, I am dealing with individual roots: numbers 1, 2, and 3. Not only must I ensure that the data from the same root connects across each of the tasks, if I screw up one step then I lose all data for that root. Second, the roots are alive, never fixed and stabilized. This means that I need to work steadily and with as compact a schedule as possible. 

On four days last week, not wanting to let the afternoons go to waste, I did a second run. I don’t think I have spent as intense a week in the lab since I was a grad student. Or maybe not even then. Starting the above about 8:30 AM, I finished around 7 PM. My wishes to be in the lab for sabbatical have come true. Is this a case of being careful for what you wish for? In my mad rush to seize the data, I have no time to analyze the data. Will it mean anything? Maybe not; but I can say for sure that if I don’t get the data then it certainly will mean nothing. Dash dash dash!