The Chinese looked at the same sky as did the other peoples of Eurasia, and, for the most part, they saw different patterns suggestive of different things. That alone is probably not surprising to scholars of the Mediterranean, Mesopotamian, Indian, and Arab worlds; what is most surprising is perhaps the sheer number of things they saw. Where Ptolemy’s (c.90–c.168 CE) star catalogue lists 1,022 stars in 48 constellations, Chen Zhuo’s 陳卓 (fl. 280–309 CE) lists 1,464 in 283 guan 官 (lit. ‘offices’), many guan consisting of but a single star or but two or three on a single line. As such, not only is the Chinese sky considerably more crowded than any other, the forms with which it is crowded are at once less distinct from one another and less visually evocative of what they represent. One would think that this would get confusing, particularly at the popular level, and one would be absolutely right. In this talk, we will examine a number of gross inconsistencies in the representation of guan ‘constellations’ in Han dynasty (206 BCE–220 CE) tombs, particularly the murals excavated at Xi’an Jiaotong University in 1987 and Qushuhao 渠樹壕 in 2015 – inconsistencies from one tomb to another and inconsistencies between the work of tomb artisans and the professional astronomical literature of the time. The interest in doing this is not to establish that tomb art falls short of ‘scientific’ standards, it is rather to identify what elements of iconography and diagrammatic representation are stable, which are not, what gets confused, and why. The sky was there for everyone to see at night, and, at the end of the day, artisans would seem to know exactly what was there, if not exactly in what order it came, of how many stars it was composed, or of what shape it actually took in the sky.