Magnetic resonance identification of an accessory submandibular duct and gland: an unusual variant J Laryngol Otol 2007;121:E18 Dear Sirs Gadodia and colleagues 1 present excellent radiological images of Bartholin's duct and the greater sublingual gland, which they have interpreted as an accessory submandibular duct and gland. This error is understandable because the detailed anatomy of the sublingual gland is much more complex than is generally appreciated. Leppi 2 discovered that the sublingual gland consists of a constant, lesser sublingual gland and a greater sublingual gland. The latter is posterior to the lesser sublingual gland in the paralingual space and was only found in 10 out of 28 people, and usually only unilaterally. Bartholin's duct runs from the greater sublingual gland to either join or open independently of Wharton's duct. The lesser sublingual gland consists of between eight to 30 small glands, from every one of which a duct of Rivinus passes to open independently on the sublingual fold. Thus, the sialograms in the paper by Gadodia and colleagues 1 show Bartholin's duct running parallel to the dilated Wharton's duct, which indicates that these ducts join, and the sublingual gland. The authors mention that there was stasis in Wharton's duct following stimulation by lemon. It would be interesting to know the situation in Bartholin's duct, where stasis would not be expected, particularly as the sublingual gland is a spontaneous secretor and continues to secrete in the absence of stimulation, and also exhibits a great resistance to obstruction.