library(magrittr)

Node <- setRefClass(
  "Node",
  fields = list(
    info = "numeric",
    left = "ANY",
    right = "ANY"
  ),
  methods = list(
    initialize = function(info) {
      .self$info <- info
      .self$left <- NULL
      .self$right <- NULL
    },
    toString = function() {
      as.character(.self$info)
    }
  )
)

BinarySearchTree <- setRefClass(
  "BinarySearchTree",
  fields = list(
    root = "ANY"
  ),
  methods = list(
    initialize = function() {
      .self$root <- NULL
    },
    create = function(val) {
      newNode <- Node$new(info = val)
      if (is.null(.self$root)) {
        .self$root <- newNode
      } else {
        current <- .self$root
        repeat {
          if (val < current$info) {
            if (!is.null(current$left)) {
              current <- current$left
            } else {
              current$left <- newNode
              break
            }
          } else if (val > current$info) {
            if (!is.null(current$right)) {
              current <- current$right
            } else {
              current$right <- newNode
              break
            }
          } else {
            break
          }
        }
      }
    }
  )
)

lca <- function(root, v1, v2) {
  if (root$info < v1 && root$info < v2) {
    lca(root$right, v1, v2)
  } else if (root$info > v1 && root$info > v2) {
    lca(root$left, v1, v2)
  } else {
    root
  }
}

stdin <- file('stdin', 'r')
inp <- readLines(stdin, n = 3, warn = FALSE)
close(stdin)

n <- as.integer(inp[1])
treeInput <- strsplit(inp[2], " ") %>% unlist %>% as.integer
vertexes <- strsplit(inp[3], " ") %>% unlist %>% as.integer
bst <- BinarySearchTree$new()
for (n in treeInput) {
  bst$create(n)
}
ans <- lca(bst$root, vertexes[1], vertexes[2])
cat(ans$info, "\n")