Mrs B was making marmalade for a competition at the Dalemain Marmalade Festival
http://www.dalemainmarmaladeawards.co.uk/ She was using a specialist "sugar thermometer" because the sugary liquid boils at well above 100 degrees Celsius. We don't tend to use mercury thermometers anymore. Digital is the order of the day. These old-fashioned ones work by having a liquid like mercury inside a narrow capillary tube. As the liquid is heated it expands and the only place it can go is up the tube. If you look at the scale you will see that the expansion is linear. The scale on the right is Celsius and that on the right is in Fahrenheit. We tend to teach that the precision is the smallest scale division - in this case + or - 2 degrees Celsius. But most physicists argue that it should be half that because you can tell where you are in the gap. We could increase the precision by making the tube narrower because there would be the same volumetric expansion with a smaller cross-sectional area. This would make the scale divisions wider and often the chance to fill in finer scale division lines. A scale like this has to be calibrated first. It is normal to fix 0 degrees Celsius with melting ice/water mixture and 100 degrees Celsius with boiling water. But only the latter appears on this scale so I wonder what they use for the second point.
