The rare earth elements (REE) are essential for a wide range of applications, from strategic assets (e.g. petroleum cracking, magnets for wind turbines) to popular merchandise, as smartphones. Since 2010, when China, the worlds close to exclusive REE supplier, imposed export quotas, several old and new deposits have been evaluated to compensate market shortage, taking advantage of significant price rises. The Araxa rare earth elements prospect boast a large reserve (6.34 Mt @5.01% RED), as well as phosphate and niobium, in a deeply weathered ore of carbonatitic origin. The mineralogy and the ore properties are unconventional for rare earth elements, and require a detailed mineralogical and technological characterisation as starting point to develop a feasible processing route. Rare earths are predominantly carried by monazite (over 70%), and by a solid solution of the plumbogummite group minerals where the barium-rich term gorceixite predominates, while cerianite and bastnaesite account for less than 1% each. Minerals of the pyrochlore supergroup are the main Nb carriers, but phosphate is also due to monazite and the plumbogummite group minerals, as apatite has barely been detected. Goethite, high-Al hematite and quartz are the main gangue minerals, and goethite is thoroughly intergrown with the other phases. Fine particle size (P-50 close to 45 mu m) and 47.4% of the REE in the -20 mu m size fraction is another feature typical of this kind of ore. The mineralogical and textural complexity of the ore required a comprehensive technological characterisation in order to evaluate processing options. Based on textural measurements, the concentration of monazite, the concentration of the REE carrying minerals and the reverse removal of quartz, as processing option for this ore, have been simulated. Incomplete liberation of monazite does limit its grade in an ideal concentrate to 80%, and its recovery to 70%. The low monazite recovery must be added to the loss of REE carried by other phases, reducing the overall REE recovery to below 45%. Monazite has also a very limited exposition of the mineral on the particle's surfaces, supposed to impair process efficiency enough to keep experimental results significantly far from the simulated ones. The concentration of the REE-bearing minerals might be efficient from the liberation point of view, and over 90% of the REE carriers can be recovered to a 97% grade concentrate. Due to the low REE grade of predominant gorceixite (3.3%), however, the concentrate's grade of 14% REE is just slightly above the double of the ore's grade. For the REE-bearing minerals taken together, the process efficiency might be hampered by selectivity due to the complex mineralogy. The major gangue minerals, goethite and hematite, are strongly intergrown with the other minerals of the assemblage, to an extent that evaluating reverse processing considering these phases was not feasible. The removal of quartz by reverse processing is quite straightforward, and 95% of the mineral might be removed to a high-grade quartz concentrate of 93%, with loss of REE of only 0.14%. The mass discharge of 8.7%, however, rises the grade of the concentrate only to 7.3% REE. Complex mineralogy and the fine crystals and particles with strong intergrowth that characterise the ore hamper efficient concentration for the Araxa REE ore, and direct hydrometallurgical processing of the whole was adopted. The results are in agreement with the few other published attempts to concentrate the rare earth minerals from residual lateritic ores related to carbonatites. (C) 2015 Elsevier B.V. All rights reserved.