Background: Mixed dentition arch analysis is an important criterion in determining an orthodontic treatment plan. The development of the Tanaka–Johnston (1974) and Moyers’ prediction (1973, 1998) was established on the Northern European population. However, the corroboration of ethnic tooth size variability suggests that prediction approaches based on a single ethnic sample may not be regarded as universal. Very few studies have been done for the Indian population. Aim and objective: The purpose of the study was done to evaluate the applicability of Tanaka–Johnston and Moyers’ mixed dentition analysis in the prediction of mesiodistal width of unerupted canines and premolars for North Indian children. Settings and design: This cross-sectional study was done on 200 participants (100 males and 100 females) in the Department of Pediatric Dentistry and Orthodontics in the North Indian population. Materials and methods: A sample of 200 North Indian population within the age group 12–15 years was randomly drawn. Mesiodistal widths of mandibular incisors and canine and premolars in both the arches were measured from the dental casts of the study participants. The sum of the actual mesiodistal widths of maxillary and mandibular canine-premolars segments was compared to those obtained from Tanaka–Johnston equations and Moyers’ prediction tables (35th to 85th percentile). Statistical analysis used: Inferential statistics were performed using unpaired and paired t-tests at a significance level of p < 0.05. Results: Moyers’ tables over-estimated the widths in maxilla and mandible of males and females at all probability levels (p < 0.001) except under-estimation in females mandibular arch only at 35% probability (p = 0.056) and at 35% and 50% probability in maxillary arch (p < 0.001 and p = 0.036, respectively). Tanaka and Johnston equations over-estimated the values in both the jaws of both the genders (p < 0.001). Conclusion: Tanaka–Johnston equations overestimated the values therefore less appropriate to be used in this population; however, Moyers’ prediction tables can be used but at different probability levels for both genders.