An experimental study of evaporating meniscus instability was performed for five capillary tube diameters ranging from 0.3 to 1.2 mm and three different volatile fluids (n-pentane, ethanol and acetone). By heating the tubes above the meniscus, the meniscus was observed to destabilize, unless the capillary tube suffered a mechanical failure due to the excess heat load. The onset of instability was recorded as a meniscus height measurement. The temperature difference across the meniscus, or superheat, at the onset of instability was calculated using the measured height and a previously validated mathematical model. Of the menisci destabilized, the superheat values varied non-linearly with capillary tube diameter. The smaller capillary tubes required higher superheats for instability. Comparison of the calculated superheats with previously developed stability criteria proved unsuccessful. An empirical relation was observed when the instability heights were scaled by the capillary geometry. Additional comparison was drawn between the capillary tubes and channels, and it was found that tubes require a larger superheat to destabilize as compared to a channel of comparable geometry. (C) 2017 Elsevier Ltd. All rights reserved.